Outcomes in Implant-based Breast Reconstruction Utilizing Biosynthetic Mesh: A Meta-analysis

Abstract

Background

Biosynthetic mesh has become increasingly popular for immediate breast cancer implant-based reconstruction as an alternative to acellular dermal matrix for soft tissue support.

Objectives

The aim of this meta-analysis was to investigate the various biosynthetic mesh options available as well as complications and outcomes.

Methods

PubMed (US National Library of Medicine, Bethesda, MD), MEDLINE (US National Library of Medicine), and Embase (Elsevier, Amsterdam, the Netherlands) were systematically reviewed for studies investigating TIGR (Novus Scientific, Uppsala, Sweden), Vicryl (Ethicon, Inc., Raritan, NJ), PDO (Poly-Med, Anderson, SC), TiLOOP (PFM Medical, Cologne, Germany), Durasorb (Integra LifeSciences, Princeton, NJ), and Galaflex (BD, Franklin Lakes, NJ) meshes, and their associated outcomes. The meta-analysis was completed in accordance with PRISMA guidelines and was performed to determine overall complication rates in patients who underwent breast reconstruction in which a mesh was used. Data were combined by a pooling of proportional outcomes as is inherent to meta-analysis. The heterogeneity of included studies was assessed in terms of Q and I² statistics.

Results

A total of 24 studies investigating 6 different types of mesh in 2167 individual breasts undergoing implant reconstruction were included. Summary effect sizes were calculated for the complications. The pooled rate of seroma formation was 5.26% (Q = 23.81%, I² = 37.01%) reported in 13 studies, hematoma formation was 2.5% (Q = 0.25%, I² = 58.27%) reported in 9 studies, skin necrosis was 5.5% (Q = 2.86%, I² = 423.78%) reported in 10 studies, infection rate was 4.8% (Q = 6.02%, I² = 149.34%) in 21 studies, and implant loss was 3.85% (Q = 6.55%, I² = 129.07%) reported in 10 studies.

Conclusions

Overall, although differences in mesh characteristics exist, the reported rate of complications is low. Biosynthetic mesh options should be taken into consideration in breast reconstruction given their demonstrated safety, significant cost advantage, and potential decrease in short-term complications in comparison to acellular dermal matrix.

Level of Evidence: 3 (Therapeutic)

Postmastectomy breast reconstruction is often pursued due to the physical and psychological benefits it offers, with up to 80% of patients receiving implant-based reconstruction vs autologous tissue flaps or no reconstruction at all.¹ Additionally, direct-to-implant reconstruction has become more popular lately compared with either delayed or 2-stage reconstruction.² Acellular dermal matrix (ADM), from human, porcine, or bovine tissue, has been the gold standard for both dual-plane submuscular reconstruction and prepectoral reconstruction to support the implant, support the soft tissue, and provide durable coverage.³ ADM materials act as a scaffold for host cells to invade and eventually create an autologous tissue layer.⁴ Although ADM use has been shown to be very effective, concerns have been expressed about higher complication rates and increased costs (up to 20 times the cost of bioabsorbable mesh).^5,6 An alternative to ADMs has recently been proposed: synthetic mesh made from absorbable or permanent fibers can be used as an internal bra for stabilizing the implant, with the added benefit of additional soft tissue support.

There are many different types of mesh, with varying characteristics, available on the market, including assorted Vicryl meshes (Ethicon, Inc., Raritan, NJ), TIGR Matrix (Novus Scientific, Uppsala, Sweden), 3DMax Mesh (BD, Franklin Lakes, NJ), polydioxanone (PDO) mesh (Poly-Med, Anderson, SC), poly-4-hydroxybutyrate (P4HB) mesh (Phasix Mesh; BD), Durasorb (Integra LifeSciences, Princeton, NJ), or Type-1a polypropylene mesh (TiLOOP Bra; PFM Medical, Cologne, Germany). Many plastic and reconstructive surgeons use either ADM or mesh in their practice due to an overall better definition of the implant pocket and inframammary fold, improved capsular contracture rates, ability to use a dual-plane approach, and less muscle dissection needed with such an approach.^4,7 Various outcomes have shown improved results with some type of soft tissue support compared with no support.^8,9

Specifically, capsular contracture rates for both ADM and synthetic meshes are significantly less than the 15% to 30% rates often cited in the literature for breast reconstruction patients without the use of any mesh.^7,10,11 More recently, rates of complications such as infection and seroma have been found to be lower in absorbable and nonabsorbable mesh groups compared with the use of ADM.^12,13 In a meta-analysis by Clark et al, biosynthetic mesh revealed a reduced risk of reoperation and explantation.¹⁴ Another potential limitation of ADM is “red breast syndrome,” an immune-mediated phenomenon in which the skin overlying ADM becomes red but has no other signs of infection, is unresponsive to antibiotics, and ultimately resolves on its own.^15,16 Red breast syndrome has been associated with many types of ADM, but most commonly with AlloDerm (LifeCell Corporation, Branchburg, NJ).^15,16 Despite the advantages of bioabsorbable mesh in breast reconstruction, these products have been developed much more recently, contributing to a delay in FDA approval for this indication.¹⁴

Given the increased popularity of synthetic meshes in postmastectomy implant reconstruction and the few recent studies showing improved outcomes, we elected to compile the available data, looking at overall complication rates and outcomes with the use of synthetic mesh alone. Given the potential advantages of synthetic mesh use, future studies should evaluate synthetic mesh and ADM by undertaking structured clinical trials. Herein, we present a systematic review of the literature to compile objective published data on the use of various synthetic meshes, and a meta-analysis of outcomes.

METHODS

This study was completed in adherence with the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines, and involved searching the PubMed (US National Library of Medicine, Bethesda, MD), MEDLINE (US National Library of Medicine), and Embase databases (Elsevier, Amsterdam, the Netherlands).¹⁷

Literature Search

The search keywords on the aforementioned databases included: “breast mesh,” “TIGR,” “TIGR mesh,” “vicryl mesh breast,” “3D max mesh” OR “3DMax,” “polydioxanone mesh,” “PDO mesh,” “poly-4-hydroxybutyrate mesh,” “P4HB mesh,” “Durasorb,” or “TiLOOP mesh” Any studies with mesh for abdominal surgery or non-English articles were excluded. We included any single-stage with immediate reconstruction or 2-stage breast reconstruction with tissue expanders with any of the following types of mesh: Vicryl, TIGR, 3DMax, PDO, P4HB, Galaflex (BD), Durasorb, or TiLOOP. Studies with a least 6 months of follow-up were included. No articles were excluded based on year of publication or language of manuscript. Two authors (A.A. and S.W.) performed independent searches to select articles that matched the inclusion criteria. Any articles that required additional screening were evaluated by 2 authors (A.A. and O.S.). Any further inclusion disputes were handled by the senior author (A.L.). The search was performed on April 20, 2024.

Data Abstraction

Authors systematically reviewed articles for any studies on breast reconstruction outcomes for patients who received implants or tissue expanders with the use of mesh support (Figure 1). Data extracted included number of patients, number of breasts, patient age, mesh type, reported follow-up, and complication rates including implant loss, infection, skin necrosis, hematoma, and seroma.

Statistics

Meta-analysis was performed to determine overall complication rates in patients who underwent breast reconstruction with the use of mesh.¹⁸ Data were combined by a pooling of proportional outcomes as inherent to meta-analysis. Hypothesis testing was performed by 2-sided Student's t-test with significance set at .05 for all arms, and a P-value was calculated for all effect summaries. Five complications were examined: hematoma, seroma, infection, skin necrosis, and implant loss. Effect summaries were recorded in Excel, and mean differences with 95% CIs for these outcomes and risk ratios with a 95% CI were calculated to estimate pooled differences across all complications. Both fixed-effects and random-effects models were used in all analyses and reported in a forest plot format. Heterogeneity of included studies was assessed in terms of the Q and I² statistics, where an I² of less than 50% is considered to be a low amount of data heterogeneity, and greater than 75% considered to be significant heterogeneity.

RESULTS

This meta-analysis includes 24 studies investigating outcomes of 6 different types of mesh in 2167 individual breasts undergoing implant reconstruction as depicted in Table 1. The types of mesh were: TIGR mesh (8 studies), Vicryl meshes (8 studies), Durasorb mesh (1 study), Galaflex (1 study), PDO mesh (3 studies), and TiLOOP mesh (3 studies).^6,9,19-39 Twelve studies involved immediate direct-to-implant breast reconstruction; only 1 study involved 2-stage reconstruction with tissue expanders and subsequent implant placement.^{9,19,21,22,24-26,28,31-33,36,37} Eight studies included both direct-to-implant and tissue expander reconstruction.^{6,19,20,23,30,34,38,39} Three studies focused on delayed breast surgery with implant placement.^27,35 The summary effect sizes for complication rates, including seroma, hematoma, skin necrosis, infection, and implant loss, are summarized in Figure 2.

The rate of seroma was reported in 13 studies, with a pooled rate of 5.26% (Q = 23.81%, I² = 37.01%, Q_v = 75.63%, $Iv2=80.17%$⁠), ranging from 1.2% (Vicryl) to 13.8% (TIGR) (Figure 3).^22,26 TIGR seroma rates averaged 5.2%, lower than the overall average of this meta-analysis.^19-26 One study reported Durasorb seroma rates of 3.7%.²⁷ The mean Vicryl seroma rate was 5.5% and the mean reported Galaflex rate was 6.0%.^6,9,28-34 PDO rates were 2.8% on average.^27,35,36 Average TiLOOP seroma rates were reported to be 10.9%.^37-39 Overall, seroma rates greatly differed between studies with the same mesh and different meshes, demonstrating significant overall heterogeneity among studies.

Postoperative hematoma was reported in 9 studies, with a pooled rate of 2.5% across all studies (Q = 0.25%, I² = 58.27%, Q_v = −5.56%, $Iv2=3.70%$⁠), ranging from 0.9% to 6.7% (Figure 4).^23,28 Casella et al report a hematoma rate of 1.37% for the TiLOOP mesh bra and Sigalove et al report a hematoma rate of 1.56% with the use of Galaflex.^34,37 TIGR mesh hematoma rates averaged 3.1%, similar to the overall hematoma rate across all included studies.^20,21,23 Mean Vicryl mesh hematoma rates of 2.7% were reported.^28-30,32 Similar to the rate of postoperative seroma, hematoma complication incidence differed significantly between and among mesh types—demonstrating an affinity to a variable-effects model (⁠$Iv2=3.70%$⁠).

Skin necrosis was reported as a complication in 10 included studies, with an overall rate of 5.5% across all breasts (Q = 2.86%, I² = 423.78%, Q_v = −47.00%, $Iv2=131.92%$⁠), ranging from 2.63% to 8.64%.^9,30 Thus, both the highest and lowest rates of skin necrosis reported were with the use of Vicryl meshes, with the mean rate of skin necrosis reported at 5.5% (Figure 5).^6,9,28,30,31 TIGR mesh skin necrosis rates averaged 5.1%, similar to the overall mean and the 6.4% skin necrosis rate for Galaflex (Sigalove).^{21,23,24,26,34}

Postoperative infection was reported in 21 studies, with a pooled complication rate of 4.8% (Q = 6.02%, I² = 149.34%, Q_v = 9.98%, $Iv2=50.32%$⁠), ranging significantly from 1.3% to 12.6% (Figure 6).^{6,9,19-21,23-39} Implant loss within the first 30 days postoperatively was also reported in 10 studies, with a pooled complication rate of 3.85% (Q = 6.55%, I² = 129.07%, Q_v = −11.75%, $Iv2=227.65%$⁠).^{6,9,19-22,24,25,28,34} The lowest reported rate of implant loss was 0.8% with the use of Durasorb (Figure 7).²⁷

DISCUSSION

ADMs have been used to improve cosmetic outcomes in breast reconstruction for almost 2 decades, since 2005, but lead to higher rates of seroma and infection than reconstruction methods without the use of ADM.^40,41 Various mesh products have been shown to be viable alternatives to ADM placement in breast reconstruction, and several studies have reported decreased complication rates and overall healthcare cost compared with ADM.^5,6 The use of mesh in breast reconstruction offers additional advantages in controlling the breast pocket, especially in submuscular implant placement. It also confers more control of cosmetic outcomes in prepectoral implant placement, often serving as additional inframammary fold support and overall implant support, in addition to implant coverage. Although various studies have compared mesh to ADM, few studies have compared outcomes among the plethora of different types of mesh available on the market. Herein, we find that outcomes are difficult to compare concertedly because methods, surgical techniques, and reporting standards are extremely variable from study to study.

There are many types of synthetic mesh available commercially for use in breast reconstruction, including completely absorbable meshes, predominantly Vicryl meshes, partially absorbable meshes (such as TIGR), and nonabsorbable meshes (such as TiLOOP), which remain in place indefinitely to support the reconstructed breast. The most common reason for choosing one type of synthetic mesh over another is usually surgeon preference, familiarity, and degree of surgeon-perceived (and often subjective) conferred cosmetic advantage. Although rates of capsular contracture were not analyzed as an outcome in this analysis, one of the primary reasons for using ADM or synthetic mesh in implant-based breast reconstruction is the significantly lower rates of capsular contracture than when placing smooth implants alone.^42,43 Surprisingly, it was found that Vicryl-only meshes were associated with greater rates of capsular contracture than ADMs but resulted in fewer reoperations for cosmetic reasons.^42,43 After reviewing this meta-analysis, surgeons should feel more comfortable using biosynthetic mesh in patients, especially those at higher risk of capsular contracture, such as patients who have had or will receive radiation, due to the low complication profile and lower rate of implant loss and reoperation.

Importantly, previous meta-analyses have shown that there is no overall difference in implant loss between human ADM, xenograft ADM, synthetic mesh, and no mesh use in implant-based breast reconstruction.⁴⁴ However, our updated review of the literature demonstrated an overall rate of implant loss at 4.1%, much lower than previously reported rates of approximately 9% across several previous studies.^45-47 Rates of implant loss did, however, vary greatly between different types of mesh, but also across studies reporting results when using the same mesh. The rates of reported implant loss ranged from 0.77% to 6.76% across the 11 studies.^{6,9,19-22,24,25,28,34} The analysis was insufficiently powered to detect statistically significant differences; however, this heterogeneity amongst reported outcomes and complications was commonly found. We believe this is likely due to significant variability in surgical technique, intraoperative sterility protocol, and patient demographics from study to study. These factors are often very difficult to objectively quantify and control unless studied within a set protocol.

Similar to the vast variability in reported implant loss, the overall seroma rate reported in our meta-analysis was 7.3%; however, this ranged from 1.1% to 7.4% amongst the included studies, with some reporting rates as high as 15%.^6,9,19-39,48 There is some evidence to suggest that thicker varieties of mesh, as well as the presence of acetone-based preservatives on the outer mesh membrane, contribute directly to increased seroma formation.⁴⁹ As such, it is difficult to compare these meshes side-by-side, even within the subgroup of Vicryl meshes, because there is no industry standard for packaging and preservative use. Even so, the pooled rates of seroma, in addition to hematoma and implant infection reported herein, despite significant heterogeneity, were still somewhat lower than those reported in the ADM literature.^{1,3,31,42,50,51} We believe that it may be difficult to demonstrate that the use of mesh vs ADM confers an advantage in postoperative complications. However, these results demonstrate, at a minimum, no significant increase, with the added benefits of superior pocket control and subjective cosmetic outcomes.

In addition to the proposed perioperative advantages and overall utility of mesh in breast reconstruction, the cost has also been demonstrated to be significantly lower than with the use of human or xenograft ADMs. Mookerjee et al report the cost per breast for Vicryl mesh at $760 total vs $9033 for an equivalent ADM.^9,33 In fact, in 2020, Faulkner et al published a study showing that, in 227 patients treated over 7 years, the cost savings achieved by using absorbable mesh instead of ADMs surpassed US$1.2 million.⁵ Thus, for some physicians, patients, and hospital systems, the use of ADMs may be cost prohibitive, prompting them to turn to absorbable mesh.

After the establishment in the United States of the federal Women's Health and Cancer Rights Act in 1998, group health insurance plans began to cover breast reconstruction after mastectomy for cancer patients.⁵² ADM is yet to be FDA-approved in the setting of breast reconstruction and is often not reimbursed or covered by insurance, making it a remarkably more expensive option than biosynthetic mesh.^14,53,54 However, biosynthetic meshes are much newer products and are thus considered as “off-label” for use in breast reconstruction by insurance companies. Their use is often not covered by insurance although biosynthetic mesh has fewer complications and is not cost prohibitive in comparison to ADM. It would be worthwhile to investigate the total cost of implant-based breast reconstruction with ADM compared to mesh, specifically investigating hospital stay and surgical/anesthesia cost in a bundled-care or episode payment model to assess total out-of-pocket expenditure and amount covered by insurance. Currently, the amount of available information regarding biosynthetic mesh and insurance coverage limits this evaluation.

Given the recency of new biological and resorbable meshes on the market, a paucity of data exists on long-term outcomes in implantable meshes. It remains unknown if long-term complications exist such as cancer-related diagnoses (eg, breast implant–associated anaplastic large cell lymphoma) that came to light with textured breast implants decades after their initial inception. Even so, concerns of long-term inflammatory responses and subsequent complications may be curbed by the significantly shorter and definitive in vivo lifespan of these mesh products in the breast. Additionally, longer-term outcomes, such as rippling, malposition, and capsular contracture, are sparsely reported in the literature and are more subjective and sporadically reported at present. Going forward, longer-term outcomes would be an important topic to investigate pending sufficient available data.

A major limitation of the meta-analysis performed was the inherent limitation of the studies available for inclusion that have been published in the literature. Furthermore, there was a large amount of heterogeneity between studies, as indicated by the large I² values, specifically in reporting rates of hematoma, infection, and implant loss. This resulted in insufficient power to detect statistically significant differences in complications between different types of mesh. In addition, many of the included studies did not segregate data amongst implant planes, ie, subpectoral, dual-plane, or subglandular/subfascia. More studies are needed that directly compare the complication profiles of different types of mesh so as to more definitively establish clinical advantages. In addition to clinical variability, differences in individual surgical technique, breast surgeon mastectomy flaps, and patient characteristics may all affect the rates of observed complications, outside of the type of mesh used. Many studies did not elaborate on the surgical technique used, or the sterility protocol. As such, with the increase in popularity of mesh use in breast reconstruction, higher-quality, standardized data are required for surgeons to make informed decisions regarding mesh type.

CONCLUSIONS

The implementation of mesh for breast reconstruction began as a method to provide an extension of the pectoralis major muscle for submuscular implant placement and for better cosmetic outcomes in prepectoral implant placement. These cosmetic benefits included lower contracture rates, and better definition and integration of breast implants. Further investigation into this topic should include increased sample sizes, patient-reported outcomes, and standardized reporting of surgical techniques and sterility protocols. Although patient-reported outcomes are not always quantifiable, inviting patient feedback could further elucidate the advantages and disadvantages of specific meshes to thereby provide a more patient-centric approach in mesh selection. Overall, surgeon experience, the cosmetic goals of the patient, product costs, and the aforementioned outcomes and complications should be taken into account when deciding which, if any, mesh should be used during breast reconstruction.

Disclosures

The authors declared no potential conflicts of interest with respect to the research, authorship, and publication of this article.

Funding

The authors received no financial support for the research, authorship, and publication of this article.

REFERENCES

Author notes