Outcome of use of acellular-dermal matrix to augment implant-based breast reconstruction
Matthew D. Barbera, Linda Williamsb, Elaine D.C. Andersona, Glyn T. Neadesa, Cameron Rainec, Oliver Younga, Dhananjay Kulkarnia, Ian Younga, J. Michael Dixona.
aEdinburgh Breast Unit, Western GeneralHospital, Edinburgh,United KingdomEH4 2XU
bCentre for Population Health Sciences, University of Edinburgh Medical School, Teviot Place, Edinburgh, United KingdomEH8 9AG
cDepartment of Plastic Surgery, St John’sHospital, Livingston, United KingdomEH54 6PP
Address for correspondence – Mr Matthew D. Barber, Consultant Breast Surgeon, Edinburgh Breast Unit, Western GeneralHospital, Edinburgh, United KingdomEH4 2XU
Tel – + 44 131 537 1629
Fax – + 44 131 537 1610
e-mail –
An original article
Presented as - “232 sheets of acellular dermal matrix or how we spent £500000 on pigskin” by Matthew D. Barber, Association of Breast Surgery Meeting, Manchester, 21st May 2013. Abstract published - Eur J Surg Oncol 2013; 39(5):517-518.
Abstract
Introduction
The use of acellular dermal matrix (ADM) has revolutionised the techniqueof implant-based breast reconstruction. It offers the option of a one-stage procedure and is felt to have benefits in cosmetic outcome but the medium and long-term outcomes are unknown.
Methods
All cases where ADM was used in a breast reconstructive procedure in the Edinburgh Breast Unit from its initial use on 7/7/2008 to 31/7/2012 were reviewed retrospectively. Follow up was completed to 30/11/2012.
Results
147 patients received 232 sheets of ADM (156 Strattice, 73 Permacol and 3 Alloderm). 40 reconstructions were lost (17.2% or 27.2% of patients). 7 of 27 (25.9%) patients requiring adjuvant therapy had this delayed due to problems with the reconstruction. 30 of 80 patients (37.5%) undergoing unilateral surgery have undergone contralateral surgery. Implant loss varied significantly with smoking (34.6% loss rate in smokers vs 13.2% in non-smokers, p=0.001), with radiotherapy (28.1% loss rate vs 13.8% with no radiotherapy, p=0.001) and with incision type. There was no statistically significant variation by operating surgeon, type of ADM used, chemotherapy use, patient weight, breast weight or nipple preservation. Patients underwent a mean of 1.54 further operations (range 0-7).
Conclusions
While offering potential cosmetic and financial benefits, the use of ADM with implant-based reconstructions has a significant rate of implant loss, further surgery and potential delay in adjuvant therapy. These must be considered when planning treatment and consenting patients.
Keywords – Breast, mastectomy, breast reconstruction, acellular dermal matrix, surgical complications.
Introduction
Implant-based breast reconstruction has tended to be less preferred to myocutaneous flaps because of limitations in cosmetic outcome, requirement for tissue expansion or a two-stage procedure and the need for ongoing maintenance. However, it offers advantages in terms of length of time taken for surgery and recovery with less interference with muscle function. The use of acellular dermal matrix (ADM) to augment implant-based reconstruction has in recent years provided the opportunity for a single-stage procedure without tissue-expansion and with suggestions of improved cosmetic outcome compared with submuscular placement, in part due to better inframammary fold definition [1-3]. Despite significant expense for the ADM, cost-savings have been suggested when compared with flap reconstructions due to reductions in theatre time and hospital stay[4]. A variety of materials have been used and include tissue derived from human, porcine and bovine sources, processed in varying manners to produce an acellular, non-reactive, connective tissue support to cover pole of the breast implant. The ADM provides a framework into which the host tissue can integrate. However, there remains very little published evidence of the outcome of this technique. Reports of inflammatory reactions to the implanted material and increased failure rates have raised concerns.
This paper presents the experience of the Edinburgh Breast Unit using ADM as part of breast reconstruction between the introduction of the technique in 2008 to 2012. The study aimed to determine the loss rate of the technique and to identify potential risk factors and implications of this.
Methods
Patients in whom ADM was used were identified from records of material ordered with cross reference to theatre lists. All cases undertaken from the first use of the technique on 7/7/2008 to 31/7/2012 were identified. Casenotes were reviewed for all patients to complete follow up to 30/11/2012. Follow up was shorter in 5 patients. One had died and 4 were discharged to distant home breast services. Cases were performed by consultant breast and plastic surgeons with experience in implant-based reconstruction or by more newly appointed consultants with experience of ADM use during training. 10 surgeons performed the primary breast and reconstructive surgery. At least initially, many cases were performed by two consultant breast surgeons or a consultant breast surgeon and a consultant plastic surgeon operating together. Decisions regarding the placement of a definitive implant (as a one-stage procedure) or tissue expander (as a one or two-stage procedure) were made by the individual surgeon with no unit protocol. Drains were placed in all cases. These were removed when drainage was less than 50mls/per day or at around 7 days. Prophylactic antibiotics were given until drains were removed. No formal assessment of cosmetic outcome was made. The recording of seroma volume and drainage was inconsistent and is therefore not presented. The occurrence of erythema in the first weeks following surgery was not formally reported but is presented where it has been noted. Adjuvant treatment was judged to have been delayed if planned treatment dates were delayed due to complications of surgery. Axillary sampling procedures included those undergoing sentinel node biopsy (61 patients) and those undergoing 4-node sampling (24 patients). Data are presented as mean (range) with median in addition for categories lacking normal distribution. Hazard ratios are presented with 95% confidence intervals. Statistical analysis was performed using univariate and multivariate Cox regression. Only those variables with a univariate p value of less than or equal to 0.2 were included in the model for multivariate analysis. Breast weight was not normally distributed and was log10 transformed prior to analysis in relation to risk of implant loss.
Results
147 patients underwent breast reconstructive surgery using ADM in 222 breasts. Mastectomies were performed to reduce risk (97), for primary cancer (84), ductal carcinoma in situ (29) and cancer recurrence (12). 3 patients have had distant cancer recurrence (one died), one patient developed breast recurrence and one an axillary recurrence. Patients had a mean age of 47.4 (range 20-75). Patients had a mean weight of 67.8kg (43-102) and BMI of 25.3 (16-43). Breast weight, reconstructed breast weight and change in breast weight are presented in table 1.
71 patients underwent bilateral reconstructions using ADM and 80 unilateral including 4 who had bilateral surgery using ADM at different times. 30 of the 80 (37.5%) patients having unilateral breast reconstruction underwent contralateral surgery, largely to improve symmetry. 232 sheets of ADM were used in total (156 Strattice (LifeCell, Bridgewater, NJ, USA), 73 Permacol (Covidien, MansfieldMA, USA) and 3 Alloderm (LifeCell, Bridgewater, NJ, USA)). In one case a sheet of Alloderm was too small to achieve implant coverage and a sheet of Strattice was used in addition. In 9 cases patients received further sheets of ADM during revisional surgery.
Mean follow up was 687 days (86-1583). 40 reconstructions were lost (17.2% of sheets of ADM involving 27.2% of patients). Loss rate was 9.5% at 3 months and 15.5% at 1 year. Losses included two instances of removal of an intact reconstruction on the contralateral side following failure on one side at the request of the patient and another patient where intact bilateral reconstructions were felt to be inadequate and were revised to latissimus dorsi flaps. Mean time to loss of implant was 167 days (9-895) with a median of 73 days. 26 reconstructed breasts were recorded as having erythema over the reconstruction in the first 4 weeks following surgery. 70 reconstructed breasts (31.5%) had problems with wound healing. 7 of 27 (25.9%) patients had a delay in adjuvant therapy as a result of problems with the reconstruction. Patients underwent 227 further operations following primary surgery with ADM (mean 1.54 per patient (range 0-7)). This included 55 planned second procedures including revision of tissue expander to implant and nipple reconstruction. Mean time to first further operation was 249 days (0-864) with a median of 208 days. 2 patients developed haematomas requiring evacuation following primary reconstructive surgery and a further 2 developed haematomas following revisional surgery. A total of 312 implants/tissue expanders were used (mean 1.4/breast (range 1-4)).
Surgeons performed surgery on between 4 and 61 breasts (mean 22.9). Implant loss rate varied from 0 to 40% between surgeons but variation was not statistically significant (p=0.51). Analysis of various factors potentially related to implant loss is presented in table 2. Implant loss rate varied significantly with smoking (p=0.002), radiotherapy to the breast/chest wall (p=0.002) and incision used (p<0.001). Loss rates were 34.6% in current smokers (hazard ratio 2.9 (1.5-5.6) versus non-smokers) and 28.1% in those who received radiotherapy. The effect of radiotherapy was primarily in those who received radiotherapy following reconstruction with a loss rate in these patients of 47.6% (hazard ratio 3.7 (1.7-7.6) versus those not receiving radiotherapy). There was no significant difference between those having had radiotherapy prior to surgery and those not receiving radiotherapy. Those with an inframammary fold incision had a loss rate of 80%. There was no statistically significant difference in loss rate with receipt of chemotherapy, ADM type, nipple preservation, use of fixed-volume implant versus expander, patient weight, breast weight, immediate reconstruction versus previous surgery (p=0.99) or indication for mastectomy (p=0.34). While there was statistically significant variation based on type of axillary surgery performed and loss rate for those undergoing axillary node clearance surgery was marked at 32.3%, this did not reach statistical significance compared with those undergoing no axillary surgery (p=0.085). Two diabetic patients underwent surgery, one of whom lost their reconstruction.
Discussion
The National Mastectomy and Breast Reconstruction Audit reported a loss rate of 8.9% for primary implant-based reconstructions at 3 months [5]. At the time of this audit few reconstructions using ADM were being performed. The loss rate in the current study of 9.5% at 3 months is not substantially different. Our findings suggest that follow up at 3 months is too early to judge the loss rate of an implant-based breast reconstructive technique as a significant number of patients can have problems after this period.
Strattice was used from August 2009 and Permacol has not been used since October 2011 and thus follow up has been longer for Permacol. However, there is no obvious difference in outcome between the two. Most published studies of the outcome of reconstructions using ADM originate from the United States and relate to use of ADMs of human origin, mainly Alloderm. Use of such products has been limited in Europe due to comparatively high cost. Numbers in these studies is variable but often low and follow up short. Published reconstruction loss rates vary from 0 to 13% [6-12]. A randomized trial of 70 patients undergoing submuscular or ADM-assisted expander/implant breast reconstruction showed no differences in the immediate post-operative period [13]. There have been no randomized studies examining results beyond this period. Two recent reviews of comparative studies of reconstructions with and without ADM (covering many of the same reports) have suggested an increased risk of failure in those in whom ADM was used with odds ratios of 2.8 and 3.0 [14, 15], although another two have shown no difference [16, 17]. Comparisons between ADM types are limited but one recent review of two surgeons using different materials found similar outcomes[18].
Reconstruction loss rates were substantially higher in women who smoked. Smoking is well known to affect wound healing and has previously been reported to substantially increase loss rates in implant-based reconstruction [19]and ADM or mesh-assisted reconstruction [12, 20]. In the present study, ex-smokers seem to lose this risk and therefore waiting several months after stopping smoking is recommended prior to performing an implant-based reconstruction with ADM.
Radiotherapy also affects wound healing and higher rates of reconstruction loss are noted [21]. Rates of reconstruction loss in those undergoing ADM-based reconstruction and receiving radiotherapy were described as “prohibitively high” by Parks et al[7]. Radiotherapy to the chest wall is routinely given to those with more than 3 lymph nodes involved by metastases. It is likely that axillary clearance surgery is a marker of those with worse disease, increasing likelihood of chest wall radiotherapy contributed to the trend towards increased loss rates with axillary clearance surgery. The receipt of chemotherapy would similarly be a marker of more advanced disease increasing likelihood of chest wall radiotherapy as well as risks of neutropaenia and infection, however, in the present study increased loss rates seen with axillary clearance surgery or chemotherapy did not reach statistical significance. Radiotherapy given prior to reconstruction did not appear to significantly increase loss rates in the present study. Similar findings have been noted previously with breast reconstruction in general [22] (however, aesthetic outcome was poorer in this study) and in those having ADM-assisted reconstruction [23]. Increased nipple necrosis has been reported in those with preoperative radiotherapy undergoing ADM or mesh-assisted reconstruction after nipple-sparing mastectomy [20].
Some surgical approaches appear to be associated with increased reconstruction loss. Those with higher risk seem to be those where scars overly the ADM. Any wound healing problem then has the potential to result in exposure of the ADM which appears not to be well tolerated. The early losses in this series with inframammary incisions are of concern but with lower and more laterally-placed incisions it is possible to close wounds over vascularised subcutaneous tissue and chest wall muscle rather than ADM. A recent study has shown a reduced rate of complications with an inframammary fold approach to mastectomy with ADM or mesh-assisted reconstruction but increased problems with periareolar incisions [20]. When skin reduction is required necessitating an inferior approach, an option is to leave a de-epithelialised layer of lower breast skin to provide further coverage of the implant. We have sometimes found it difficult to achieve implant coverage with de-epithelialised skin alone and thus often use this approach in combination with ADM. It remains to be seen whether these variations improve loss rates.
While there were large variations in patient and breast weight, these did not affect loss of reconstruction in the present study. Implant-based reconstruction failure rates have previously been reported to be substantially increased with obesity [19]. Two large reviews of outcome after expander and ADM-augmented implant reconstruction suggested a small but statistically significant increase in risk of complications with body mass index with a hazard ratio of around 1.1 [20, 24].
There is a feeling that use of a tissue expander may result in less skin tension following surgery and hence reduce wound healing problems compared with a fixed volume implant (although this may result in the need for a two-stage approach). However, the present study shows no difference in loss rates between expanders and implants. While nipple preservation was not a significant risk factor for implant loss, the replacement of the nipple as a graft did seem to be associated with a high loss rate and is no longer performed on our unit. There were insufficient diabetic patients in the current study to offer a definitive judgement on risk but caution appears warranted in this group.
Implant-based breast reconstruction with ADM is associated with a high rate of reconstruction loss, delay in adjuvant treatment for a significant number of patients and high rates of revisional and contralateral surgery. It does offer the possibility of a cosmetically satisfactory one-stage reconstruction but discussion with patients should cover risks of implant loss and likelihood of further surgery in detail. Careful consideration should be given as to whether this reconstructive approach is suitable for women in whom there is an expectation that adjuvant radiotherapy or chemotherapy will be recommended and it should be used with great caution in smokers.
Conflict of interest statement
The authors have no conflict of interest regarding this study.
Funding
No funding was received for this study.
References
1. Vardanian AJ, Clayton JL, Roostaeian J, Shirvanian V, Da Lio A, Lipa JE, Crisera C, Festekjian JH. Comparison of implant-based immediate breast reconstruction with and without acellular dermal matrix. Plast Reconstr Surg 2011; 128(5):403e-410e.
2. Nguyen KT, Mioton LM, Smetona JT, Seth AK, Kim JY. Esthetic Outcomes of ADM-Assisted Expander-Implant Breast Reconstruction. Eplasty 2012; 12:e58.
3. Forsberg CG, Kelly DA, Wood BC, Mastrangelo SL, Defranzo AJ, Thompson JT, David LR, Marks MW. Aesthetic outcomes of acellular dermal matrix in tissue expander/implant-based breast reconstruction. Ann Plast Surg 2013; epub
4. Johnson RK, Wright CK, Gandhi A, Charny MC, Barr L. Cost minimisation analysis of using acellular dermal matrix (Strattice™) for breast reconstruction compared with standard techniques. Eur J Surg Oncol 2013; 39(3):242-7.
5. NHS Information Centre. National Mastectomy and Breast Reconstruction Audit Third Report 2010. [accessed 5 October 2013]
6. Salzberg CA, Dunavant C, Nocera N. Immediate breast reconstruction using porcine acellular dermal matrix (Strattice™): long-term outcomes and complications. J Plast Reconstr Aesthet Surg 2013; 66(3):323-8.
7 Parks JW, Hammond SE, Walsh WA, Adams RL, Chandler RG, Luce EA. Human acellular dermis versus no acellular dermis in tissue expansion breast reconstruction. Plast Reconstr Surg 2012; 130(4):739-46.
8. Venturi ML, Mesbahi AN, Boehmler JH 4th, Marrogi AJ. Evaluating sterile human acellular dermal matrix in immediate expander-based breast reconstruction: a multicenter, prospective, cohort study. Plast Reconstr Surg 2013; 131(1):9e-18e.
9. Colwell AS, Damjanovic B, Zahedi B, Medford-Davis L, Hertl C, Austen WG Jr. Retrospective review of 331 consecutive immediate single-stage implant reconstructions with acellular dermal matrix: indications, complications, trends, and costs. Plast Reconstr Surg 2011; 128(6):1170-8.
10. Collis GN, TerKonda SP, Waldorf JC, Perdikis G. Acellular dermal matrix slings in tissue expander breast reconstruction: are there substantial benefits? Ann Plast Surg 2012; 68(5):425-8.