Acellular dermal matrix reduces capsule formation in two‐stage breast reconstruction^{, ,}

Abstract

Acellular dermal matrix (ADM) is commonly employed to create an inferior pocket for the tissue expander in two‐stage breast reconstruction. The authors sought to determine whether placement of ADM during the first stage of reconstruction decreases the amount of capsule formation at implant exchange. Patients who underwent mastectomy and tissue expander reconstruction were included in this study. Two biopsies were obtained at the time of implant exchange, one from the pocket adjacent to the ADM and the other from the area adjacent to the pectoralis muscle. Pathology analysis was performed on each sample. Ten patients underwent immediate breast reconstruction with Alloderm during the 3‐month study period. Capsule thickness was significantly greater in the areas where the expander was in direct contact with the pectoralis muscle (782 ± 194 µm) compared to those in contact with human acellular dermal matrix (hADM) (47·91 ± 110·82 µm; P < 0·05). Analysis of the sub‐pectoral capsule demonstrated diffuse deposition of collagen, neutrophils, contractile myofibroblasts and synovia‐like metaplasia, characteristic of a foreign body response. Conversely, within the inferior pocket where the hADM was in direct contact with the expander, we noted migration of host epithelial cells, fibroblasts, mesenchymal cells and angiogenesis, indicating host tissue regeneration. Acellular dermal matrix, when placed at the first stage of breast reconstruction, significantly reduces thickness and inflammatory character of the capsule in comparison to the patient's native tissue.

Introduction

Implant‐based breast reconstruction is increasingly sought by patients following mastectomy. Rates increased by an average of 11% per year between 1998 and 2008, overtaking autologous reconstruction as the primary reconstructive modality in this population and accounts for more than one third of all breast reconstructions 1. Two‐stage immediate implant‐based reconstruction involves the creation of a subpectoral pocket into which a tissue expander is placed. Serial expansions augment the pocket, allowing subsequent exchange for a permanent implant. However, complications associated with two‐stage TE/implant‐based breast reconstruction are common. Capsular contracture is a highly unpredictable, but frequently encountered, complication, with reported rates as high 29% in the reconstructed breast 2. Clinically, capsular contracture may present as pain, firmness of the breast and implant migration with consequent aesthetic deformation of the reconstructed breast, often requiring reoperation to correct the problem 3. At the tissue level, capsule formation demonstrates the hallmarks of a classic foreign body reaction: deposition of collagen bundles and elastin, the presence of inflammatory cells, fibroblasts and synovial‐like metaplasia 4. Active contractile elements of the capsule cause progressive concentric tightening, producing the clinical sequelae of capsular contracture observed.

Although a variety of pathophysiological mechanisms underlying capsular contracture have been proposed, the general consensus involves a long‐standing subclinical inflammatory process. A study by Siggelkow et al. evaluated the capsules of 53 breast implants in 43 patients in an attempt to correlate histological changes to the Baker–Spear degree of capsular contracture 5. Their work found that an increasing degree of capsular contracture was related to increased overall histological evidence of inflammation, synovial‐like metaplasia and capsule thickness. In an effort to mitigate inflammation and subsequent capsular contracture in implant‐based breast reconstruction and to provide additional support to the TE/implant, Breuing and Warren first described the creation of an inferolateral sling using Alloderm (LifeCell Corporation, Branchburg, NJ), a cadaveric human acellular dermal matrix (hADM) 6. This technique has proven a valuable alternative to total submuscular coverage in prosthetic breast reconstruction and has gained wide, although not universal, acceptance since its initial description 6, 7. In a recent regional survey of ASPS surgeons, nearly three quarters acknowledged using ADM for breast reconstruction 7. The purported benefits of ADM use include better inferior support and coverage, more control over implant and inframammary fold position, decreased pain because of and potentially decreased incidence of capsular contracture 6, 8.

A growing body of literature, including both animal and clinical studies, has documented decreased inflammatory reaction and reduced capsule formation in integrated hADM versus native tissue 9, 10, 11. In light of these findings, we sought to evaluate and compare the histological characteristics of the capsule formed on the area of the breast with the hADM sling with that of the native submuscular breast capsule. We hypothesised that decreased markers of inflammation and capsular thickness would be observed in portions of the capsule lining hADM versus native submuscular pocket (Figures 1 and 2).

Figure 1.

Superior pocket with visible capsule as demonstrated by the ‘pearl‐white’ appearance. The biopsy site is clearly demarcated (black box).

Figure 2.

Inferior pocket with human acellular dermal matrix (hADM), which has fully incorporated at this stage. The biopsy site is clearly demarcated (black box).

Methods

Approval was granted by the University of Maryland School of Medicine Institutional Review Board (IRB) for the conduction of this study. Women who underwent mastectomy with immediate two‐stage tissue expander/implant‐based reconstruction with hADM inferior sling between January and April 2012 were eligible for inclusion in the study. Exclusion criteria included all patients undergoing immediate single‐stage implant‐based reconstruction or reconstruction with autologous tissue. In total, 10 patients (19 breasts) underwent two‐stage reconstruction with tissue expander/implants; 9 patients underwent bilateral mastectomy and reconstruction, while a single patient had unilateral mastectomy and reconstruction. All surgeries were performed by surgeons in the Division of Plastic Surgery at the University of Maryland Medical Center.

At the time of mastectomy, reconstruction was carried out in a manner similar to previously described methods 6. In brief, a single 10 × 16 cm sheet of hADM (LifeCell Corporation) was cut to the contour of the inframmary fold. The inferior border of pectoralis major was elevated to create a subpectoral pocket. The cut hADM was sutured inferiorly to the inframammary fold and laterally to the serratus anterior fascia to create an inferior sling. The tissue expander was then place in the subpectoral pocket, and the inferior border of the pectoralis major was sutured to the superior edge of the hADM sheet. The expander was expanded as much as the overlying skin flap would tolerate so as to ensure adequate obliteration of dead space but would not cause pressure necrosis of overlying tissue.

After mastectomy and tissue expander placement, eligible patients were invited to participate in the study. After a period of approximately 4–6 months, permitting time for completion of chemotherapeutic regimens, patients underwent exchange of the tissue expander for a permanent silicone implant. The mastectomy scar was excised, and capsulotomies were performed, permitting access to the tissue expander. The tissue expander was removed, and rhomboid‐shaped biopsies, roughly 2·0 × 1·0 cm, were sharply excised. Biopsies were taken from two locations of each breast capsule – the superior portion of the pocket directly overlying the pectoralis major and the inferior portion of the pocket lining the hADM sling (Figures 1 and 2). Biopsy samples were fixed in formalin and sent for histological analysis (Tejas Pathology, Tomball, Texas). Formalin‐fixed, paraffin‐embedded sections of the excised capsule were hematoxylin and eosin (H&E)‐stained, and capsular thickness was assessed. This measurement was defined as the width between the capsule surface at the interface with the breast pocket and underlying identifiable tissue or bioprosthetic material.

Results

A total of 10 patients (19 breasts) underwent immediate two‐stage breast reconstruction using tissue expanders and permanent silicone implants following mastectomy. Of these patients, nine had bilateral reconstruction, while one patient underwent unilateral reconstruction only. The average age of the subjects included in the study was 49·1 years. The average body mass index (BMI) was 28·6. Three plastic surgeons performed all of the reconstructions included in this analysis. The surgeons utilised identical techniques for the first‐stage reconstruction using Alloderm^® (LifeCell Corp.) hADM and tissue expander placement. Sientra tissue expanders (Santa Barbara, CA, USA) were used in all cases. Indications for mastectomy and breast reconstruction included breast cancer Stage I (n = 3), Stage II (n = 4), Stage III (n = 1) and Ductal Carcinoma in Situ DCIS (n = 1). The breast cancer staging information was unavailable for one patient. In addition, four patients received radiation to the affected breast prior to staged reconstruction, and a total of six patients underwent either adjuvant (n = 3) or neoadjuvant (n = 3) chemotherapy. The average time from expander placement to implant exchange and capsule biopsy was 158 days (Table 1).

Table 1.

Baseline clinical/demographic variables

Total patients	10
Average age (years)	49·1 ± 14·1
Average BMI (kg/m2)	28·6 ± 7·7
Total reconstructed breasts	19
Unilateral reconstructions	1
Bilateral reconstructions	9
Breast cancer
Stage I	3 (30%)
Stage II	4 (40%)
Stage III	1 (10%)
DCIS	1 (10%)
Number of patients irradiated	4
Before implant	4 (100%)
After implant	0 (0%)
Chemotherapy
None	3 (33%)
Neoadjuvant	3 (33%)
Adjuvant	3 (33%)

BMI, body mass index.

At the time of tissue expander exchange and placement of permanent silicone implant, the superior breast capsule was easily identifiable grossly because of its characteristic pearly, avascular appearance within the expander pocket (Figure 1). The inferior aspect of the presumed capsule that lined the hADM sling had a grossly distinct appearance, lacking the pearly quality of the superior capsule (Figure 2). While the borders of the Alloderm were clearly distinguishable, the material appeared grossly to have extensively integrated into the adjacent host tissue with near‐complete epithelialisation and neovascularisation. Resultant bleeding occurred upon sharply incising the integrated Alloderm for biopsy.

Superior capsule biopsies overlying the pectoralis major muscle demonstrated substantial collagen deposition at the pocket interface as well as the presence of neutrophil and myofibroblast infiltration, suggestive of active inflammation (Figure 3). A layer of synovial‐like metaplasia was also observed in these sections (Figure 4).

Figure 3.

Capsule formation (C) overlying the pectoralis after prolonged contact with the breast expander. Diffuse infiltration of neutrophils (round purple nuclei) can be seen, indicative of active inflammation.

Figure 4.

Capsule formation (C) with the unique overlying synovial layer (SL). Areas of neutrophil infiltration (solid arrows) are seen at the pectoralis (M)/capsule (C) interface.

Histological analysis of the inferior aspect of the pocket lining the implanted hADM demonstrated host epithelial cell migration into the hADM with concomitant fibroblast, mesenchymal cell and vascular integration (Figures 5 and 6). Minimal to non‐existent collagen deposition was observed in the portion of the capsule lining hADM, and no discernible synovial‐like metaplasia was found. Incomplete integration of the ADM into the native tissue was observed in a subset of slides from some patients. The presence of inflammatory cells in the newly remodelled matrix was not observed, and no evidence of infection was present in any sections. Capsular thickness was found to be significantly greater in the superior capsule biopsies superficial to pectoralis compared with inferior capsule lining hADM (782 ± 194 versus 47·91 ± 110 µm, respectively; P < 0·05) (Figure 7).

Figure 5.

Alloderm with early migration of fibroblasts into the matrix. Blood vessels (solid arrows) can also be seen deep into the matrix, which is important for matrix integration and remodelling.

Figure 6.

Late Alloderm integration demonstrating integration with more extensive infiltration of host fibroblasts (open arrows) and blood vessels.

Figure 7.

Capsule thickness in superior biopsies adjacent to pectoralis muscle comparted to inferior biopsies adjacent to human acellular dermal matrix (hADM).

Discussion

Capsular contracture is a significant problem in implant‐based breast reconstruction. In this study, we aimed to evaluate the histological characteristics of the tissue formed around the breast tissue expander device in the region of the ADM as compared to the native submuscular breast capsule that is not in contact with the ADM. We hypothesised that there would be decreased capsular thickness and overall decreased inflammation in the region of the breast pocket containing the ADM as indicated by the presence of inflammatory markers and an absence of the synovial‐like metaplasia seen in the native breast pocket. To do so, we conducted a prospective trial and enrolled 10 patients undergoing a total of 19 two‐staged implant‐based breast reconstructions (Figures 6 and 7).

In the native pocket, where host tissues are in direct contact with the tissue expander, we found significantly thicker capsule formation, as defined by extensive collagen deposition and the presence of myofibroblasts. In addition, we noted increased neutrophil and fibroblast infiltration into the native capsule. The hADM‐lined portion of the pocket had a much thinner capsular thickness and with minimal collagen deposition and inflammatory cell infiltration as demonstrated by pathology and laboratory analysis. These results were consistent with previously published data from our group, demonstrating that implanted devices that were partially wrapped in ADM resulted in less capsule formation than those devices that were not wrapped in ADMs in a primate model 10.

Researchers have studied the cellular mechanisms involved in capsular formation and capsular contracture. Studies analysing targeted inhibitory therapeutic modalities have included agents such as angiotensin‐converting enzyme inhibitors in rabbit models 12 and the leukotriene inhibitor zafirlukast in rat models 13. Further studies, such as the one by our group, studied capsule formation around implants with and without hADM in a primate model. H&E staining showed a lack of capsule in the hADM‐covered specimens. Immunostaining with antibodies to alpha‐smooth muscle actin highlighted that there was an abundance of these contractile elements in the non‐hADM group but were minimal in the hADM‐coated implants. However, at that time, our follow‐up was only 10 weeks. Another study by Basu et al. evaluated the inflammatory response and foreign body reaction seen when hADM is used during implant‐based breast reconstruction. At the time of implant exchange, two biopsies were taken. The first biopsy was from the native subpectoral capsule, and the other biopsy from the inferior pocket with the integrated ADM. Upon immunohistological analysis, specimens derived from ADM showed reduced signs of inflammation, with statistically significantly decreased granulation tissue, chronic inflammation, capsule fibrosis and fibroblast cellularity, and foreign body giant cell inflammatory reaction as compared to the internal control from the native breast capsule. The results of this study are consistent with the results of Basu et al. and further corroborate the idea that at least on a cellular and histological level, the use of ADM reduces capsule formation, and when capsule formation does occur, it is less inflammatory and thus perhaps less likely to become reactive and lead to contracture. It is also apparent that there is a difference in surfaces when the implant is surrounded by ADM versus native muscle only. The implant–muscle interface compared to the hADM–implant interface results in increased movement and tension and, therefore, increased mechanotransduction. Mechanotransduction, or the reaction of cells to biomechanical stimuli, is known to result in increased collagen, myofibroblast stimulation and inflammation 14, 15, findings consistent with our results. Further contributing to this phenomenon are immunoreactants within the muscle, which may react differently when exposed to hADM, implant alone or a combination of the two.

The literature on secondary or tertiary revision breast surgery has several reports of reduced capsular contracture with the use of ADM. Spear et al. reported managing 42 breasts with Baker stage III/IV capsular contracture after implant‐based breast reconstruction with capsulectomy and the use of hADM with an 88% success rate 16. Nahabedian et al. reviewed their personal techniques utilising ADM in secondary breast reconstruction and reported good early success after capsulectomy and use of ADM by placing it over the device and suturing to the perimeter of the breast pocket and surrounding tissues 17. Similarly, Maxwell et al. reported on 78 consecutive patients who underwent revisionary breast augmentation/mastopexies with ADM. One hundred and nineteen patients had capsular contracture, and at a minimum of 6 months follow‐up, no patients had clinically significant capsular contracture. Maxwell et al. proposes that the mechanism by which ADM can help prevent capsular contracture has to do with creating an environment that promotes healing without excess inflammation 18.

Although Maxwell et al. did not publish their results of animal studies at various time points to demonstrate decreased inflammation at those time points within the breast pocket, our data does support their hypothesis and is overall consistent with the idea that the use of ADM may act to decrease capsule formation and therefore capsular contracture 18.

While the results of our study are promising regarding reduced capsule formation associated with hADM use in two‐stage implant‐based reconstruction, several limitations should be noted. First, our study was designed only to assess the effect of hADM on the histology of capsule formation from the time of tissue expander placement and implant exchange, a follow‐up time of roughly 5 months. This short time frame does not allow observation of the long‐term histological changes that occur in the capsule lining the hADM sling. Capsule formation and deposition of contractile elements leading to capsular contracture undoubtedly continue beyond our investigational period and likely contribute to late capsular contracture. Furthermore, although we and others have demonstrated that integrated HADM reduces the thickness of lining capsule, no correlation between capsular thickness and severity of capsular contracture has been definitively demonstrated. Thus, we cannot necessarily extrapolate the results of this study to expected clinical findings. The sample size is also small, which limits our ability to make conclusions, especially if there is an effect on capsule formation in the radiation subset. However, because all samples were taken from the same breast pocket, the study is neatly and internally normalised; any factor such as radiation or infection or haematoma should theoretically affect both groups similarly.

Conclusion

This study supports the growing body of literature demonstrating reduced inflammation and subsequent capsule formation with the use of hADM in breast reconstruction. Based on a limited sample, we have reasonably good evidence that hADM use results in a thinner capsule and one that contains fewer inflammatory histological components in comparison to the native subpectoral pocket when used to create inferior support for breast reconstruction.