- Email: [email protected]
Breast fat grafting (lipomodelling) after extended latissimus dorsi flap breast reconstruction: A preliminary report of 200 consecutive cases
Journal of Plastic, Reconstructive & Aesthetic Surgery (2010) 63, 1769e1777
Breast fat grafting (lipomodelling) after extended latissimus dorsi flap breast reconstruction: A preliminary report of 200 consecutive cases ´bastien Garson, Thomas Delaporte, Raphael Sinna, Emmanuel Delay*, Se Gilles Toussoun University of Lyon, Department of Plastic and Reconstructive Surgery, Le´on Be´rard Cancer Centre, 28 rue Laennec, 69008 Lyon, France Received 31 January 2009; accepted 1 December 2009
KEYWORDS Lipomodelling; Fat grafting; Breast reconstruction; Extended latissimus dorsi; Breast; Fat transfer
Summary Background: The efficacy of fat grafting has long been a controversial issue. Breast lipomodelling after extended latissimus dorsi flap reconstruction was first attempted at the Plastic and Reconstructive Surgery unit of Leon Berard Cancer Centre in 1999. We present the results of a retrospective report of the first 200 consecutive patients treated at our institution from 1999 to 2003. Methods: We identified specific requirements of the patients, and collected information on the surgical techniques used and the volumes of fat tissue injected. We analysed and compared the results of a total of 244 lipomodelling sessions. Results: The graft consisted of 70% fat graft, 13% oily supernatant and 17% serum residues. Approximately 30% was lost during centrifugation. On average, 176 ml of fat were injected in each breast. Very satisfactory results were obtained in 94.5% of the cases, with a majority of patients (80%) being very satisfied with the procedure and only 1.5% complications. Conclusion: Our results demonstrate the safety and feasibility of breast lipomodelling. It is a new approach to improve reconstructive outcome after extended latissimus flap breast reconstruction. ª 2009 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved.
Because of the immediate, medium-term and long-term benefits associated with autologous latissimus dorsi flap, we often use this technique in breast reconstruction.1e4 In certain cases (thin patients or after secondary muscle
atrophy), the final volume of the breast may be found insufficient. In these patients, the preferred strategy is the secondary insertion of an underlying implant, but the procedure is no longer autologous and the patients may
* Corresponding author. Tel.: þ33 478 78 27 67. E-mail address: [email protected] (E. Delay). 1748-6815/$ - see front matter ª 2009 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.bjps.2009.12.002
1770
R. Sinna et al.
suffer specific side effects associated with prosthesis placement. There was no approach allowing to improve the results of primary autologous latissimus dorsi flap reconstruction while preserving the autologous nature of the procedure. Authors such as Coleman have shown that careful grafting using a rigorous standardised technique allows to preserve the fat graft5e8 and that the grafted tissues do not deteriorate over time at the graft site.6,9,10 Therefore, we started fat grafting in 1998 in breast reconstruction. At that time, fat grafting to the breast was highly controversial among plastic surgeons. The disagreement, which had arisen after the description by Bircoll, in 1987, of cosmetic breast augmentation using autologous fat and liposuction techniques,11 was temporarily settled by the American Society of Plastic and Reconstructive Surgery (ASPRS).12 The very satisfactory results obtained in facial rejuvenation have encouraged us to adapt the technique for use in autologous latissimus dorsi flap breast reconstruction. We showed that lipomodelling helps achieve satisfactory contouring and texture, as well as natural-looking cleavage in patients undergoing latissimus dorsi flap breast reconstruction. We also assess the impact of fat injections on breast radiological images, using mammography, ultrasound and magnetic resonance imaging. The preliminary results of our study confirmed that lipomodelling causes no specific interpretation difficulties for radiologists.13 The aim of the present study was to describe the preliminary report of 200 consecutive patients undergoing latissimus dorsi flap and lipomodelling breast reconstruction without an implant.
Patients and methods We retrospectively reviewed the first 200 consecutive patients treated with lipomodelling in our institution. All patients had been operated by the same surgeon (E. Delay) for the primary reconstruction and the lipomodelling session.
Preoperative work-up Careful clinical examination of the breast is required and the areas to be treated must be marked before the surgery (Figure 1)a and b. In our department, three-dimensional morphological images, as well as the usual two-dimensional images, are taken and compared to assess the quantity of fatty tissue to be transferred and estimate fat resorption (Figure 2). The various fatty areas in the patient’s body are examined to identify and locate natural fat deposits (e.g., flanks, inner thighs, buttock, abdomen and lateral thighs).
Figure 1 ab. Preoperative site marking on standard photograph. Preoperative site marking is done by the surgeon to identify harvest and graft sites, breast areas involved in contralateral breast symmetrisation or nipple-areola reconstruction.
follow-up (second or third) sessions can be performed under local anaesthesia.
Fat harvest Fat is harvested using a disposable 15-cm blunt-tipped cannula (3 mm in diameter, Thiebaud-Biomedical Instruments). There is no infiltration.
The lipomodelling technique This technique has been described previously.14,15 Harvesting fat graft from the abdominal area facilitates patient positioning. Because lipomodelling is usually combined with other procedures (either liposuction of the inframammary fold, reconstruction of the nippleeareola complex or contralateral symmetrisation), it is generally performed under general anaesthesia. Less extensive lipomodelling or
Figure 2 Preoperative digitisation.
volume
assessment
with
3D
Breast fat grafting (lipomodelling) after extended latissimus dorsi flap breast reconstruction
1771
Fat graft is collected with a 10-ml Luer Lock syringe fitted directly on to the cannula (Figure 3). A 3-min centrifugation at 3200 rpm allows to obtain fat in three layers (Figure 4)a and b. Only the intermediate layer contains purified fat graft suitable for transplantation. The other layers are discarded.
Fat transfer When all fat grafts have been processed, a number of 10-ml syringes of purified fat aspirates are available. Small incisions in the breast are made with a 17-gauge trocar, which gives sufficient access while the minimal, punctate residual scars will be practically invisible. Several incisions are made around the breast and through old scars, so that the area can be honeycombed with numerous microtunnels for the fat transfer. Fat is injected to the breast by adapting specific disposable cannulas of 2 mm in diameter. These cannulas are slightly longer and more resistant than those used for fat transfer to the face since the mechanical constraints are increased when injecting more solid and more fibrous tissues. Fat grafts are injected from deep to superficial tissues, while gradually withdrawing the cannula with a crossing fan-shape infiltration with multiple passes into multiple tunnels in all layers that one can inject. The fat is injected in all layers: subcutaneous tissues, pectoral muscle and latissimus dorsi flap. The volume of fat to be transferred (volume assessed preoperatively and subjectively to achieve symmetry) to the patient must be overestimated to account for a 30% fat resorption after injection, in our experience. When the receptor tissues are so saturated (when injected fat is coming out from the incision) with fat that they can absorb no more, there is no point in continuing injections, or areas of fat necrosis might arise. It is preferable to perform a follow-up session after a few months; the procedure will be much easier and will achieve better results. Sutures to the breast are done using extra-fine absorbable stitches,
Figure 4 a- Syringes before centrifugation. b- Syringes after centrifugation, with the 3 layers: oil (top), purified fat cells (intermediate), blood residues (bottom).
and an ordinary dry bandage is applied and left in place for a few days.
Clinical assessment of patients Clinical data collected from the medical records of the first 200 patients with autologous latissimus dorsi flap breast reconstruction and lipomodelling operated on by the senior author were retrospectively reviewed until November 2005. All patients were operated between 1999 and 2003. The main data collected were:
Figure 3 incision.
Fat harvest with a syringe through a periumbilical
- Dates of mastectomy and of first consultation - Date of reconstruction by latissimus dorsi flap and of different lipomodelling procedure - The main medical history, including radiotherapy, hormonotherapy and chemotherapy and tobacco use - Weight and height - For the immediate breast reconstruction the main surgical specificity, especially the weight of the mastectomy - The different element of the lipomodelling procedure - Type of anaesthesia, operating duration
1772 Table 1
R. Sinna et al. Complementary treatment, and type of reconstruction for the 200 patients
nb of pers.
Tobacco
Chemotherapy
Hormonotherapy
Radiotherapy
Total
Immediate breast reconstruction Delayed breast reconstruction Other
2 20 0
6 90 1
5 77 1
13 128 1
53 146 1
Total
22
97
83
142
200
- Volume of fat harvested, prepared and injected - Area from where the fat was removed - Different surgical procedures that were done at the same time - Immediate and late outcomes For the purpose of statistical analysis, cases of bilateral breast reconstruction were randomised (considering only either the left or the right breast) and recorded as unilateral. A descriptive analysis of all collected data was done; then we assessed the influence of the different factors on lipomodelling (e.g., radiotherapy, tobacco use, chemotherapy and localisation of fat harvesting). The morphological characteristics of the patients were assessed by two surgeons and one secretary at each patient visit. Only data collected at last visit were taken into account. Morphological features were evaluated on a 4-point Likert scale ‘very good’, ‘good’, ‘medium’ and ‘poor’. Patient satisfaction was assessed at each visit and rated on a 4-point Likert scale ‘very satisfied’, ‘satisfied’, ‘partially satisfied’ and ‘dissatisfied’.
Results Descriptive analysis Of the 200 patients included in the study, 145 (73%) had delayed breast reconstruction (LBR) and 54 (26.5%) immediate breast reconstruction (EBR) after breast cancer surgery. One patient (0.5%) underwent reconstruction of the breast and the upper thorax for management of sequelae after surgical treatment of an Ewing sarcoma of the second rib. There was a slight predominance of left breast reconstruction (53.5%) over right breast (46.5%). Fourteen patients (7%) had bilateral involvement. The median age at time of lipomodelling was 48.7 years (range, 18e72 years). The median duration of follow-up for the 200 patients was 14.5 months. (range, 4e52 months). The patients were offered lipomodelling an average of 11.7 months (range, 4 monthse6 years) after the latissimus dorsi reconstruction.
Table 2
A review of patient records revealed that 11% had a history of tobacco use and 71% had received radiotherapy to the breast area. This rate increased to 88.3% in the subgroup of patients undergoing delayed breast reconstruction (Table 1). The median body mass index (BMI) of the patients was 23.1 (range, 17e35.5) with a median body weight of 60.7 kg and a median height of 162 cm. No important weight variations of the patients were observed between latissimus dorsi flap reconstruction and the first lipomodelling session. The BMI at the first lipomodelling session was 23 (range 16.7e37.7); the patients lost an average of 270 g between the two procedures. Lipomodelling In total, 244 lipomodelling sessions were performed. A total of 37 patients had at least two sessions and seven patients had three sessions of lipomodelling. Therefore, the average number of sessions per patient was 1.22; of these, 1.01 were performed under general anaesthesia. The number and distribution of general versus local anaesthesia in the 200 patients studied are given in Table 2. Cells were collected by liposuction of the abdomen in 189 patients (94.5%), of the buttocks in eight (4%) or of the trochanter in three (1.5%). The mean duration of the sessions (second procedure of breast reconstruction) including breast symmetrisation, complementary liposuction, nippleeareola complex reconstruction and lipomodelling, was 102 min (ranging from 60 to 150 min). The morphological indication for lipomodelling was insufficient breast projection and lack of fullness in 96% of the women, with correction of local deformities in 18%. The mean volume of fat collected at the first session was 276 ml (range, 60e635 ml). There was a 30% volume loss during centrifugation (ratio: 0.7094) and 194 ml of fat were left available for injection. On average, harvested tissues contained 49 ml of oil and 65 ml of serum. After centrifugation and preparation, the harvested fat contained 70% fat, 13% oil and 17% serum residues. Only the ‘pure’ fat is injected; oil and serum are discarded. The mean volume of fat injected in each breast was 176 ml (range, 35e405 ml).
Type of anaesthesia for each session General anaesthesia
1st session 2nd session 3rd session
27% 73% 1%
Local anaesthesia
Total percentage (200 patients)
Number of patients
Percentage
Number of patients
Percentage
175 26 1
87.5% 13% 0.5%
25 11 6
12.5% 5.5% 3%
100% 18.5% 3.5%
Breast fat grafting (lipomodelling) after extended latissimus dorsi flap breast reconstruction Table 3 Procedures that were performed during the lipomodelling session Fre ´quence Pourcentage Symetrisation controlateral breast Tattooing Nipple reconstruction Dog ear correction Inframammary fold fixation Liposuction of the inframammary fold
135 patients 164 136 33 15 98
(67.5%) 82% 68% 16,5% 7,5% 49%
In total, 135 patients (67.5%) underwent symmetrisation, with a mean reduction mammaplasty of 157 g (range, 10e 770 g). Several other surgical procedures were associated with lipomodelling: corrective tattooing nipple reconstruction, dog-ear correction, inframammary fold fixation and liposuction of the inframammary fold (Table 3). Among the 200 patients, only 23 patients underwent exclusively a lipomodelling procedure. Surgical fixation of the inframammary fold was necessary when the fold had been disrupted by abdominal advancement flap reconstruction.1 Liposuction was used to emphasise the inframammary fold and eliminate possible excesses of fat in the armpit area or in the lower inner quadrant. Second and third lipomodelling follow-up sessions were not combined with other surgical procedures. The statistical analysis of factors likely to influence fat cell transfer did not reach significance. Although the
1773
number of non-abdominal liposuctions is too low for accurate interpretation, fat graft collected from the trochanter or the buttocks seemed to produce a higher yield than abdominal fat, with less than 20% cell loss after centrifugation. Improved skin trophicity and colour were reported following lipomodelling, but no precise objective evaluation was available. Complications Complications were reported in three patients: two cases of minor local infection, easily controlled by antibiotics and by removal of the suture corresponding to the redness, and one case of pneumothorax requiring pleural drainage. Five patients had clinical signs of cytosteatonecrosis, though with no serious morphological or diagnostic consequences. They presented as a palpable nodule. They were all surgically removed. Evaluation of results Of the 200 breast reconstructions analysed, nine (4.5%) were rated by the clinical team as satisfactory and 191 (95.5%) as very satisfactory. No medium or poor results were reported. The study showed that patients were satisfied (20%) or very satisfied (80%) with the results. None expressed dissatisfaction. The secondary cosmetic benefit derived from the liposuction also increased patient satisfaction with the procedure (Figures 5e7).
Figure 5 Second line of breast reconstruction in a patient undergoing delayed reconstruction with latissimus dorsi flap and abdominal advancement flap. a. full-face preoperative view, b. three-quarter preoperative view, c. full-face postoperative view at 1 year, d. three-quarter postoperative view at 1 year.
1774
R. Sinna et al.
Figure 6 Second line of breast reconstruction in a patient undergoing delayed reconstruction with latissimus dorsi flap and abdominal advancement flap. a. full-face preoperative view, b. three-quarter preoperative view, c. full-face postoperative view at 1 year, d. three-quarter postoperative view at 1 year.
Figure 7 Second line of breast reconstruction in a patient undergoing immediate reconstruction with latissimus dorsi flap. a. fullface preoperative view, b. three-quarter preoperative view, c. full-face postoperative view at 1 year, d. three-quarter postoperative view at 1 year.
Breast fat grafting (lipomodelling) after extended latissimus dorsi flap breast reconstruction
Discussion The main objective of surgeons who specialise in breast reconstruction is to create a breast with natural-like shape and texture and in perfect symmetry with the contralateral breast. Autologous procedures prevent the risk of complications associated with the presence of a foreign body as prostheses. Results remain stable over time.9,10,16 The shape of the breast reconstructed with autologous tissues is much like the contralateral breast and does not affect the body image of the patient, especially as the patient generally recovers deep cutaneous sensibility.2 To meet the needs and demands of today’s patients, plastic surgeons keep striving for better results. In this context, lipomodelling appears as a very attractive tool; it allows to shape the breast on demand and to correct deformities that could occur after first-stage reconstructive surgery. Besides, using the fat graft from the patient preserves the autologous nature of the reconstruction procedure. The concept of fat transfer is not new. Neuber is thought to have conducted the first experiment more than a century ago.17 In fact, the first attempt at breast reconstruction by injection of fat was reported in 1895 by Czerny who transplanted a giant lipoma to a woman with postoperative breast deformities.18 Recent studies in the field5e8 have shown that, when handled with care and following strict technical guidelines, fatty tissue grafts are like any other graft and remain stable over the long term.6,9,10,19 By contrast, Guerrerosantos et al. have shown that the type of recipient site is critical for the survival of the graft, with lower resorption rates when fat graft are grafted into muscle.20 The autologous latissimus dorsi flap happens to be a suitable recipient tissue for fat cell injections. Following our presentation at the 2001 meeting of the SOFCPRE (French society of plastic reconstructive and aesthetic surgery), some of our colleagues argued that injecting fat graft into the muscle might cause fat embolism; however, this complication was neither observed in the 200 women of the current series nor in our 880 patients treated to date. Our concept of breast reconstruction does not use the latissimus dorsi flap as a ‘skin carrier’ to bring to the chest wall enough skin to cover a prothesis. With the abdominal advancement flap reconstruction,1 the latissimus dorsi flap is placed under the skin only to restore the breast mound and to bring adequate volume for further breast lipomodelling. The grafted flap promotes the vascularisation of injected fat graft. Autologous latissimus dorsi flap reconstruction is therefore the first step of the reconstruction procedure, with lipomodelling being used as the second, finishing step. There are many advantages to using lipomodelling. Fat graft are nearly always at hand and easily available though liposuction. Most patients have sufficient fat tissues available for harvest. The lack of available fat graft has never been an obstacle to lipomodelling in our patients undergoing latissimus dorsi flap reconstruction (94.5% of patients had sufficient abdominal fat). The surgical technique is simple, well understood and accepted by the patients, but requires specific training of the surgeons. As the surgeons
1775
acquire more experience, they achieve better results, notably with larger volumes of fat transferred. The technique is now well established, thereby ensuring reproducible results. When the results are judged insufficient, more sessions can be scheduled, which is generally well accepted by the patients since postoperative requirements are kept to a minimum. The long-term stability of the graft remains controversial, especially as there is no objective tool for calculating breast volumes.7 Using magnetic resonance imaging, Hoerl and Feller21 have shown that fat retention stabilises after 6 months. With a median follow-up of 14.5 months in our 200 patients, we have observed that lipomodelling achieves stable long-term improvement and that volume gains can be considered permanent after 6 months. We are undergoing a to-be-published objective study on fat resorption using a 3D tool. A significant secondary cosmetic benefit is associated with the removal of unsightly fat pads at the site of harvest. Most patients and surgeons are satisfied with the principle of removing fat in the abdominal region and the thighs to improve breast contouring. The slimming benefit achieved by liposuction most certainly enhances the overall satisfaction of the patients with the procedure, especially when repeated sessions are required. This indirect benefit most likely accounts for the remarkably high satisfaction of the patients with lipomodelling. This is no wonder because when the results of the first session are judged insufficient, the patients can undergo a second, or even a third session, until complete satisfaction is achieved. Fat grafting seems to increase neovascularisation resulting in local skin improvement,11,22 which is of particular interest for irradiated patients. The mechanism of fat graft survival is not clear, and the role of adiposederived stem cells and preadipocytes in fat survival remains to be determined.23,24 A correct technique, as described (fan shape, multitunnel, etc.) could avoid clinical oil cyst to form. Our five cases of cystoteatonecrosis appeared at the beginning of our experience. Therefore, a short training period is required for surgeons performing lipomodelling. They must be able to transfer important volumes of fat. The rate of fat resorption is approximately and subjectively, around 30%, in our experience. Therefore, the graft must represent at least 140% of the target volume. Besides, with only 70% of pure fat graft obtained after centrifugation, the volume harvested must be approximately 200% of the target volume. This implies harvesting many syringes of fat, which takes time. To keep the duration of the procedure to a minimum, it is advisable to perform the different steps simultaneously: the surgeon harvests the fat graft, then a surgical assistant prepares the graft for centrifugation and injection while the surgeon proceeds with further steps of the procedure, such as breast symmetrisation. In agreement with previous reports,25e28 it seems that the buttocks and the trochanter are the highest yield sites, with more than 80% fat graft. However, these results remain controversial.6,28 By contrast, harvesting fat from these sites implies that the patient be moved, which makes the procedure longer. Whenever possible, we therefore prefer to harvest fat graft from the abdomen.
1776 Major postoperative complications of lipomodelling are rare.22,23,29,30 As for liposuction, patients complain of pain and oedema. Only three (1.5%) major complications were reported in our series of 200 patients, with two infections at the graft site, which were easily controlled without compromising patient outcome or delaying recovery, and one case of pneumothorax requiring simple pleural drainage. The most serious drawback in our series was graft resorption. The rates of resorption and survival of the fat graft grafted to the patients are hardly predictable. We subjectively used rates around 30%, whereas other reports have used values between 30% and 80%.7 We are currently developing an evaluation system for the objective assessment of fat resorption after lipomodelling.31,32 This 3D surface scan system provides a 3D model of the patient. Comparing images taken before and after the procedure makes it possible to calculate volume differences and find out the rate of fat resorption. Preliminary results indicate rates between 30% and 40%, according to the patient, reduced by approximately 20% at the second lipomodelling session.
Future prospects Lipomodelling is a useful complement to autologous latissimus dorsi flap reconstruction of the breast. Using this procedure, it has been possible to extend the indications for flap reconstruction in our patients (95%) and also to improve the quality of breast reconstruction.14 Currently, we propose the latissimus dorsi flap to only those patients with enough fat on the dorsal thoracic area. Now, with the second stage of lipomodelling, we can complete the lack of volume during the second surgery and also propose this technique to more patients. The principle of latissimus dorsi flap reconstruction has been modified and the autologous tissue flap is now used as a recipient site for further fat cell injections. Preliminary radiological studies have demonstrated the absence of deleterious consequences for breast radiological follow-up, thus permitting to extend the indications for breast lipomodelling.13,33e35 We have tested the procedure in patients requiring only partial breast reconstruction, or in those with breast deformities (pectus excavatum and Poland’s syndrome), tuberous breasts, unilateral hypotrophy or asymmetric breasts.14,15,36,37 From the results obtained in these patients, it seems reasonable to extend the use of lipomodelling to women seeking cosmetic breast surgery.19 From the present study, we conclude that breast lipomodelling is safe and effective for patients undergoing breast reconstruction with autologous latissimus dorsi flap. Grafting fat graft to the reconstructed breast helps improve the very satisfactory results obtained with primary plastic surgery, which makes autologous flap reconstruction the treatment of choice for our patients. Using fat graft taken from the patient helps preserve the autologous nature of the procedure. Autologous latissimus dorsi flap placement can be seen as the first step of the reconstruction, which prepares the ground for further enhancement using lipomodelling. Lipomodelling is simple, safe and reproducible, which makes it available for use in many indications of breast
R. Sinna et al. reconstructive and plastic surgery. The technique thus appears as one of the most considerable advances in breast reconstructive surgery since the mid-1990s.
Conflict of interest None.
Funding source None.
References 1. Delay E, Jorquera F, Pasi P, et al. Autologous latissimus breast reconstruction in association with abdominal advancement flap: a new refinement in breast reconstruction. Ann. Plast. Surg 1999;42:67e75. 2. Delay E, Jorquera F, Lucas R, et al. Sensitivity of breasts reconstructed with the autologous latissimus dorsi flap. Plast. Reconstr. Surg 2000;106:302e9. 3. Delay E, Gounot N, Bouillot A, et al. Breast reconstruction with the autologous latissimus dorsi flap. Preliminary report of 60 consecutive reconstructions. Ann. Chir. Plast. Esthet 1997;42: 118e30. 4. Delay E, Delaporte T, Sinna R. Breast implant alternatives. Ann. Chir. Plast. Esthet 2005;50:652e72. 5. Coleman SR. Long-term survival of fat transplants: controlled demonstrations. Aesthetic Plast. Surg 1995;19:421e5. 6. Chajchir A. Fat injection: long-term follow-Up. Aesthetic Plast. Surg 1996;20:291e6. 7. Sinna R, Delay E, Garson S, et al. Scientific bases of fat transfer. Critical review of the literature. Ann. Chir. Plast. Esthet 2006;51:223e30. 8. Coleman SR. Facial recontouring with lipostructure. Clin Plast Surg 1997;24:347e67. 9. Missana MC, Laurent I, Barreau L, et al. Autologous fat transfer in reconstructive breast surgery: indications, technique and results. Eur J Surg Oncol 2007;33:685e90. 10. Spear SL, Wilson HB, Lockwood MD. Fat injection to correct contour deformities in the reconstructed breast. Plast Reconstr Surg 2005;116:1300e5. 11. Bircoll M. Cosmetic breast augmentation utilizing autologous fat and liposuction techniques. Plast. Reconstr. Surg 1987;79: 267e71. 12. ASPRS Ad-Hoc Committee on New Procedures. Report on autologous fat transplantation, September 30, 1987. 13. Pierrefeu-lagrange AC, Delay E, Guerin N, et al. Radiological evaluation of breasts reconstructed with lipomodelling. Ann. Chir. Plast. Esthet 2006;51:18e28. 14. Delay E. Lipomodelling of the reconstructed breast. In: Spear SL, editor. Surgery of the breast: principles and art. Philadelphia, Pa: Lippincott-Raven Publishers; 2006. 15. Delaporte T, Delay E, Toussoun G, et al. Breast volume reconstruction by lipomodeling technique: about 15 consecutive cases. Ann. Chir. Plast. Esthet 2009;54:303e16. 16. Clough KB, Louis-Sylvestre C, Fitoussi A, et al. Donor site sequelae after autologous breast reconstruction with an extended latissimus dorsi flap. Plast. Reconstr. Surg 2002;109: 1904e11. 17. Neuber F. Fetttransplantation. Chir kongr vehandl dtsch ges chir 1893;22:66. 18. Czerny V. Plasticher erzatz der Brustdruse durch ein lipom. Zentralb chir 1895;27:72.
Breast fat grafting (lipomodelling) after extended latissimus dorsi flap breast reconstruction 19. Zheng DN, Li QF, Lei H, et al. Autologous fat grafting to the breast for cosmetic enhancement: experience in 66 patients with long-term follow-up. J Plast Reconstr Aesthet Surg 2008; 61:792e8. 20. Guerrerosantos J, Gonzalez-Mendoza A, Masmela Y, et al. Long-term survival of free fat grafts in muscle: an experimental study in rats. Aesthetic Plast. Surg 1996;20:403e8. 21. Hoerl HW, Feller AM In: Shiffman MA, editor. Autologous fat volume retention: evaluation by magnetic resonance imaging, in autologous fat transplantation. New York: Marcel Dekker, Inc; 2001. p. 31e42. 22. Chajchir A, Benzaquen I. Fat-grafting injection for soft-tissue augmentation. Plast. Reconstr. Surg 1989;84:921e34. 23. Coleman SR. Structural fat grafting: more than a permanent filler. Plast Reconstr Surg 2006;118:108Se120S. 24. Rigotti G, Marchi A, Galie ´ M, et al. Clinical treatment of radiotherapy tissue damage by lipoaspirate transplant: a healing process mediated by adipose-derived adult stem cells. Plast Reconstr Surg 2007;119:1409e22. 25. Pinski KS, Roenigk Jr HH. Autologous fat transplantation. Longterm follow-up. J Dermatol. Surg. Oncol 1992;18:179e84. 26. Asken S. Autologous fat transplantation: micro and macro techniques. Am J Cosm Surg 1987;4:111. 27. Hudson DA, Lambert EV, Bloch CE. Site selection for fat autotransplantation: some observations. Aesthetic Plast. Surg 1990;14:195e7. 28. Rohrich RJ, Sorokin ES, Brown SA. In search of improved fat transfer viability: a quantitative analysis of the role of centrifugation and harvest site. Plast. Reconstr. Surg. 2004;113:391e5.
1777
29. Illouz YG. Present results of fat injection. Aesthetic Plast. Surg 1998;12:175e81. 30. De Pedroza LV. Fat transplantation to the buttocks and legs for aesthetic enhancement or correction of deformities: long-term results of large volumes of fat transplant. Dermatol. Surg 2000; 26:1145e9. 31. Garson S, Delay E, Sinna R, et al. 3D evaluation and breast plastic surgery: preliminary study. Ann. Chir. Plast. Esthet 2005;50:296e308. 32. Sinna R, Garson S, Taha F, et al. Evaluation of 3D numerisation with structured light projection in breast surgery. Ann. Chir. Plast. Esthet 2009;50:317e30. 33. Amar O, Bruant-Rodier C, Lehmann S, et al. Fat tissue transplant: restoration of the mammary volume after conservative treatment of breast cancers, clinical and radiological considerations. Ann Chir Plast Esthet 2008;53:169e77. 34. Carvajal J, Patin ˜o JH. Mammographic findings after breast augmentation with autologous fat injection. Aesthet Surg J 2008;28:153e62. 35. Gosset J, Guerin N, Toussoun G, et al. Radiological evaluation after lipomodelling for correction of breast conservative treatment sequelae. Delay E. Ann Chir Plast Esthet 2008 Apr; 53:178e89. 36. Pinsolle V, Chichery A, Grolleau JL, et al. Autologous fat injection in Poland’s syndrome. J Plast Reconstr Aesthet Surg 2008;61:784e91. 37. Delay E, Gosset J, Toussoun G, et al. Efficacy of lipomodelling for the management of sequelae of breast cancer conservative treatment. M. Ann Chir Plast Esthet 2007;53:153e68.
Breast fat grafting (lipomodelling) after extended latissimus dorsi flap breast reconstruction: A preliminary report of 200 consecutive cases ´bastien Garson, Thomas Delaporte, Raphael Sinna, Emmanuel Delay*, Se Gilles Toussoun University of Lyon, Department of Plastic and Reconstructive Surgery, Le´on Be´rard Cancer Centre, 28 rue Laennec, 69008 Lyon, France Received 31 January 2009; accepted 1 December 2009
KEYWORDS Lipomodelling; Fat grafting; Breast reconstruction; Extended latissimus dorsi; Breast; Fat transfer
Summary Background: The efficacy of fat grafting has long been a controversial issue. Breast lipomodelling after extended latissimus dorsi flap reconstruction was first attempted at the Plastic and Reconstructive Surgery unit of Leon Berard Cancer Centre in 1999. We present the results of a retrospective report of the first 200 consecutive patients treated at our institution from 1999 to 2003. Methods: We identified specific requirements of the patients, and collected information on the surgical techniques used and the volumes of fat tissue injected. We analysed and compared the results of a total of 244 lipomodelling sessions. Results: The graft consisted of 70% fat graft, 13% oily supernatant and 17% serum residues. Approximately 30% was lost during centrifugation. On average, 176 ml of fat were injected in each breast. Very satisfactory results were obtained in 94.5% of the cases, with a majority of patients (80%) being very satisfied with the procedure and only 1.5% complications. Conclusion: Our results demonstrate the safety and feasibility of breast lipomodelling. It is a new approach to improve reconstructive outcome after extended latissimus flap breast reconstruction. ª 2009 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved.
Because of the immediate, medium-term and long-term benefits associated with autologous latissimus dorsi flap, we often use this technique in breast reconstruction.1e4 In certain cases (thin patients or after secondary muscle
atrophy), the final volume of the breast may be found insufficient. In these patients, the preferred strategy is the secondary insertion of an underlying implant, but the procedure is no longer autologous and the patients may
* Corresponding author. Tel.: þ33 478 78 27 67. E-mail address: [email protected] (E. Delay). 1748-6815/$ - see front matter ª 2009 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.bjps.2009.12.002
1770
R. Sinna et al.
suffer specific side effects associated with prosthesis placement. There was no approach allowing to improve the results of primary autologous latissimus dorsi flap reconstruction while preserving the autologous nature of the procedure. Authors such as Coleman have shown that careful grafting using a rigorous standardised technique allows to preserve the fat graft5e8 and that the grafted tissues do not deteriorate over time at the graft site.6,9,10 Therefore, we started fat grafting in 1998 in breast reconstruction. At that time, fat grafting to the breast was highly controversial among plastic surgeons. The disagreement, which had arisen after the description by Bircoll, in 1987, of cosmetic breast augmentation using autologous fat and liposuction techniques,11 was temporarily settled by the American Society of Plastic and Reconstructive Surgery (ASPRS).12 The very satisfactory results obtained in facial rejuvenation have encouraged us to adapt the technique for use in autologous latissimus dorsi flap breast reconstruction. We showed that lipomodelling helps achieve satisfactory contouring and texture, as well as natural-looking cleavage in patients undergoing latissimus dorsi flap breast reconstruction. We also assess the impact of fat injections on breast radiological images, using mammography, ultrasound and magnetic resonance imaging. The preliminary results of our study confirmed that lipomodelling causes no specific interpretation difficulties for radiologists.13 The aim of the present study was to describe the preliminary report of 200 consecutive patients undergoing latissimus dorsi flap and lipomodelling breast reconstruction without an implant.
Patients and methods We retrospectively reviewed the first 200 consecutive patients treated with lipomodelling in our institution. All patients had been operated by the same surgeon (E. Delay) for the primary reconstruction and the lipomodelling session.
Preoperative work-up Careful clinical examination of the breast is required and the areas to be treated must be marked before the surgery (Figure 1)a and b. In our department, three-dimensional morphological images, as well as the usual two-dimensional images, are taken and compared to assess the quantity of fatty tissue to be transferred and estimate fat resorption (Figure 2). The various fatty areas in the patient’s body are examined to identify and locate natural fat deposits (e.g., flanks, inner thighs, buttock, abdomen and lateral thighs).
Figure 1 ab. Preoperative site marking on standard photograph. Preoperative site marking is done by the surgeon to identify harvest and graft sites, breast areas involved in contralateral breast symmetrisation or nipple-areola reconstruction.
follow-up (second or third) sessions can be performed under local anaesthesia.
Fat harvest Fat is harvested using a disposable 15-cm blunt-tipped cannula (3 mm in diameter, Thiebaud-Biomedical Instruments). There is no infiltration.
The lipomodelling technique This technique has been described previously.14,15 Harvesting fat graft from the abdominal area facilitates patient positioning. Because lipomodelling is usually combined with other procedures (either liposuction of the inframammary fold, reconstruction of the nippleeareola complex or contralateral symmetrisation), it is generally performed under general anaesthesia. Less extensive lipomodelling or
Figure 2 Preoperative digitisation.
volume
assessment
with
3D
Breast fat grafting (lipomodelling) after extended latissimus dorsi flap breast reconstruction
1771
Fat graft is collected with a 10-ml Luer Lock syringe fitted directly on to the cannula (Figure 3). A 3-min centrifugation at 3200 rpm allows to obtain fat in three layers (Figure 4)a and b. Only the intermediate layer contains purified fat graft suitable for transplantation. The other layers are discarded.
Fat transfer When all fat grafts have been processed, a number of 10-ml syringes of purified fat aspirates are available. Small incisions in the breast are made with a 17-gauge trocar, which gives sufficient access while the minimal, punctate residual scars will be practically invisible. Several incisions are made around the breast and through old scars, so that the area can be honeycombed with numerous microtunnels for the fat transfer. Fat is injected to the breast by adapting specific disposable cannulas of 2 mm in diameter. These cannulas are slightly longer and more resistant than those used for fat transfer to the face since the mechanical constraints are increased when injecting more solid and more fibrous tissues. Fat grafts are injected from deep to superficial tissues, while gradually withdrawing the cannula with a crossing fan-shape infiltration with multiple passes into multiple tunnels in all layers that one can inject. The fat is injected in all layers: subcutaneous tissues, pectoral muscle and latissimus dorsi flap. The volume of fat to be transferred (volume assessed preoperatively and subjectively to achieve symmetry) to the patient must be overestimated to account for a 30% fat resorption after injection, in our experience. When the receptor tissues are so saturated (when injected fat is coming out from the incision) with fat that they can absorb no more, there is no point in continuing injections, or areas of fat necrosis might arise. It is preferable to perform a follow-up session after a few months; the procedure will be much easier and will achieve better results. Sutures to the breast are done using extra-fine absorbable stitches,
Figure 4 a- Syringes before centrifugation. b- Syringes after centrifugation, with the 3 layers: oil (top), purified fat cells (intermediate), blood residues (bottom).
and an ordinary dry bandage is applied and left in place for a few days.
Clinical assessment of patients Clinical data collected from the medical records of the first 200 patients with autologous latissimus dorsi flap breast reconstruction and lipomodelling operated on by the senior author were retrospectively reviewed until November 2005. All patients were operated between 1999 and 2003. The main data collected were:
Figure 3 incision.
Fat harvest with a syringe through a periumbilical
- Dates of mastectomy and of first consultation - Date of reconstruction by latissimus dorsi flap and of different lipomodelling procedure - The main medical history, including radiotherapy, hormonotherapy and chemotherapy and tobacco use - Weight and height - For the immediate breast reconstruction the main surgical specificity, especially the weight of the mastectomy - The different element of the lipomodelling procedure - Type of anaesthesia, operating duration
1772 Table 1
R. Sinna et al. Complementary treatment, and type of reconstruction for the 200 patients
nb of pers.
Tobacco
Chemotherapy
Hormonotherapy
Radiotherapy
Total
Immediate breast reconstruction Delayed breast reconstruction Other
2 20 0
6 90 1
5 77 1
13 128 1
53 146 1
Total
22
97
83
142
200
- Volume of fat harvested, prepared and injected - Area from where the fat was removed - Different surgical procedures that were done at the same time - Immediate and late outcomes For the purpose of statistical analysis, cases of bilateral breast reconstruction were randomised (considering only either the left or the right breast) and recorded as unilateral. A descriptive analysis of all collected data was done; then we assessed the influence of the different factors on lipomodelling (e.g., radiotherapy, tobacco use, chemotherapy and localisation of fat harvesting). The morphological characteristics of the patients were assessed by two surgeons and one secretary at each patient visit. Only data collected at last visit were taken into account. Morphological features were evaluated on a 4-point Likert scale ‘very good’, ‘good’, ‘medium’ and ‘poor’. Patient satisfaction was assessed at each visit and rated on a 4-point Likert scale ‘very satisfied’, ‘satisfied’, ‘partially satisfied’ and ‘dissatisfied’.
Results Descriptive analysis Of the 200 patients included in the study, 145 (73%) had delayed breast reconstruction (LBR) and 54 (26.5%) immediate breast reconstruction (EBR) after breast cancer surgery. One patient (0.5%) underwent reconstruction of the breast and the upper thorax for management of sequelae after surgical treatment of an Ewing sarcoma of the second rib. There was a slight predominance of left breast reconstruction (53.5%) over right breast (46.5%). Fourteen patients (7%) had bilateral involvement. The median age at time of lipomodelling was 48.7 years (range, 18e72 years). The median duration of follow-up for the 200 patients was 14.5 months. (range, 4e52 months). The patients were offered lipomodelling an average of 11.7 months (range, 4 monthse6 years) after the latissimus dorsi reconstruction.
Table 2
A review of patient records revealed that 11% had a history of tobacco use and 71% had received radiotherapy to the breast area. This rate increased to 88.3% in the subgroup of patients undergoing delayed breast reconstruction (Table 1). The median body mass index (BMI) of the patients was 23.1 (range, 17e35.5) with a median body weight of 60.7 kg and a median height of 162 cm. No important weight variations of the patients were observed between latissimus dorsi flap reconstruction and the first lipomodelling session. The BMI at the first lipomodelling session was 23 (range 16.7e37.7); the patients lost an average of 270 g between the two procedures. Lipomodelling In total, 244 lipomodelling sessions were performed. A total of 37 patients had at least two sessions and seven patients had three sessions of lipomodelling. Therefore, the average number of sessions per patient was 1.22; of these, 1.01 were performed under general anaesthesia. The number and distribution of general versus local anaesthesia in the 200 patients studied are given in Table 2. Cells were collected by liposuction of the abdomen in 189 patients (94.5%), of the buttocks in eight (4%) or of the trochanter in three (1.5%). The mean duration of the sessions (second procedure of breast reconstruction) including breast symmetrisation, complementary liposuction, nippleeareola complex reconstruction and lipomodelling, was 102 min (ranging from 60 to 150 min). The morphological indication for lipomodelling was insufficient breast projection and lack of fullness in 96% of the women, with correction of local deformities in 18%. The mean volume of fat collected at the first session was 276 ml (range, 60e635 ml). There was a 30% volume loss during centrifugation (ratio: 0.7094) and 194 ml of fat were left available for injection. On average, harvested tissues contained 49 ml of oil and 65 ml of serum. After centrifugation and preparation, the harvested fat contained 70% fat, 13% oil and 17% serum residues. Only the ‘pure’ fat is injected; oil and serum are discarded. The mean volume of fat injected in each breast was 176 ml (range, 35e405 ml).
Type of anaesthesia for each session General anaesthesia
1st session 2nd session 3rd session
27% 73% 1%
Local anaesthesia
Total percentage (200 patients)
Number of patients
Percentage
Number of patients
Percentage
175 26 1
87.5% 13% 0.5%
25 11 6
12.5% 5.5% 3%
100% 18.5% 3.5%
Breast fat grafting (lipomodelling) after extended latissimus dorsi flap breast reconstruction Table 3 Procedures that were performed during the lipomodelling session Fre ´quence Pourcentage Symetrisation controlateral breast Tattooing Nipple reconstruction Dog ear correction Inframammary fold fixation Liposuction of the inframammary fold
135 patients 164 136 33 15 98
(67.5%) 82% 68% 16,5% 7,5% 49%
In total, 135 patients (67.5%) underwent symmetrisation, with a mean reduction mammaplasty of 157 g (range, 10e 770 g). Several other surgical procedures were associated with lipomodelling: corrective tattooing nipple reconstruction, dog-ear correction, inframammary fold fixation and liposuction of the inframammary fold (Table 3). Among the 200 patients, only 23 patients underwent exclusively a lipomodelling procedure. Surgical fixation of the inframammary fold was necessary when the fold had been disrupted by abdominal advancement flap reconstruction.1 Liposuction was used to emphasise the inframammary fold and eliminate possible excesses of fat in the armpit area or in the lower inner quadrant. Second and third lipomodelling follow-up sessions were not combined with other surgical procedures. The statistical analysis of factors likely to influence fat cell transfer did not reach significance. Although the
1773
number of non-abdominal liposuctions is too low for accurate interpretation, fat graft collected from the trochanter or the buttocks seemed to produce a higher yield than abdominal fat, with less than 20% cell loss after centrifugation. Improved skin trophicity and colour were reported following lipomodelling, but no precise objective evaluation was available. Complications Complications were reported in three patients: two cases of minor local infection, easily controlled by antibiotics and by removal of the suture corresponding to the redness, and one case of pneumothorax requiring pleural drainage. Five patients had clinical signs of cytosteatonecrosis, though with no serious morphological or diagnostic consequences. They presented as a palpable nodule. They were all surgically removed. Evaluation of results Of the 200 breast reconstructions analysed, nine (4.5%) were rated by the clinical team as satisfactory and 191 (95.5%) as very satisfactory. No medium or poor results were reported. The study showed that patients were satisfied (20%) or very satisfied (80%) with the results. None expressed dissatisfaction. The secondary cosmetic benefit derived from the liposuction also increased patient satisfaction with the procedure (Figures 5e7).
Figure 5 Second line of breast reconstruction in a patient undergoing delayed reconstruction with latissimus dorsi flap and abdominal advancement flap. a. full-face preoperative view, b. three-quarter preoperative view, c. full-face postoperative view at 1 year, d. three-quarter postoperative view at 1 year.
1774
R. Sinna et al.
Figure 6 Second line of breast reconstruction in a patient undergoing delayed reconstruction with latissimus dorsi flap and abdominal advancement flap. a. full-face preoperative view, b. three-quarter preoperative view, c. full-face postoperative view at 1 year, d. three-quarter postoperative view at 1 year.
Figure 7 Second line of breast reconstruction in a patient undergoing immediate reconstruction with latissimus dorsi flap. a. fullface preoperative view, b. three-quarter preoperative view, c. full-face postoperative view at 1 year, d. three-quarter postoperative view at 1 year.
Breast fat grafting (lipomodelling) after extended latissimus dorsi flap breast reconstruction
Discussion The main objective of surgeons who specialise in breast reconstruction is to create a breast with natural-like shape and texture and in perfect symmetry with the contralateral breast. Autologous procedures prevent the risk of complications associated with the presence of a foreign body as prostheses. Results remain stable over time.9,10,16 The shape of the breast reconstructed with autologous tissues is much like the contralateral breast and does not affect the body image of the patient, especially as the patient generally recovers deep cutaneous sensibility.2 To meet the needs and demands of today’s patients, plastic surgeons keep striving for better results. In this context, lipomodelling appears as a very attractive tool; it allows to shape the breast on demand and to correct deformities that could occur after first-stage reconstructive surgery. Besides, using the fat graft from the patient preserves the autologous nature of the reconstruction procedure. The concept of fat transfer is not new. Neuber is thought to have conducted the first experiment more than a century ago.17 In fact, the first attempt at breast reconstruction by injection of fat was reported in 1895 by Czerny who transplanted a giant lipoma to a woman with postoperative breast deformities.18 Recent studies in the field5e8 have shown that, when handled with care and following strict technical guidelines, fatty tissue grafts are like any other graft and remain stable over the long term.6,9,10,19 By contrast, Guerrerosantos et al. have shown that the type of recipient site is critical for the survival of the graft, with lower resorption rates when fat graft are grafted into muscle.20 The autologous latissimus dorsi flap happens to be a suitable recipient tissue for fat cell injections. Following our presentation at the 2001 meeting of the SOFCPRE (French society of plastic reconstructive and aesthetic surgery), some of our colleagues argued that injecting fat graft into the muscle might cause fat embolism; however, this complication was neither observed in the 200 women of the current series nor in our 880 patients treated to date. Our concept of breast reconstruction does not use the latissimus dorsi flap as a ‘skin carrier’ to bring to the chest wall enough skin to cover a prothesis. With the abdominal advancement flap reconstruction,1 the latissimus dorsi flap is placed under the skin only to restore the breast mound and to bring adequate volume for further breast lipomodelling. The grafted flap promotes the vascularisation of injected fat graft. Autologous latissimus dorsi flap reconstruction is therefore the first step of the reconstruction procedure, with lipomodelling being used as the second, finishing step. There are many advantages to using lipomodelling. Fat graft are nearly always at hand and easily available though liposuction. Most patients have sufficient fat tissues available for harvest. The lack of available fat graft has never been an obstacle to lipomodelling in our patients undergoing latissimus dorsi flap reconstruction (94.5% of patients had sufficient abdominal fat). The surgical technique is simple, well understood and accepted by the patients, but requires specific training of the surgeons. As the surgeons
1775
acquire more experience, they achieve better results, notably with larger volumes of fat transferred. The technique is now well established, thereby ensuring reproducible results. When the results are judged insufficient, more sessions can be scheduled, which is generally well accepted by the patients since postoperative requirements are kept to a minimum. The long-term stability of the graft remains controversial, especially as there is no objective tool for calculating breast volumes.7 Using magnetic resonance imaging, Hoerl and Feller21 have shown that fat retention stabilises after 6 months. With a median follow-up of 14.5 months in our 200 patients, we have observed that lipomodelling achieves stable long-term improvement and that volume gains can be considered permanent after 6 months. We are undergoing a to-be-published objective study on fat resorption using a 3D tool. A significant secondary cosmetic benefit is associated with the removal of unsightly fat pads at the site of harvest. Most patients and surgeons are satisfied with the principle of removing fat in the abdominal region and the thighs to improve breast contouring. The slimming benefit achieved by liposuction most certainly enhances the overall satisfaction of the patients with the procedure, especially when repeated sessions are required. This indirect benefit most likely accounts for the remarkably high satisfaction of the patients with lipomodelling. This is no wonder because when the results of the first session are judged insufficient, the patients can undergo a second, or even a third session, until complete satisfaction is achieved. Fat grafting seems to increase neovascularisation resulting in local skin improvement,11,22 which is of particular interest for irradiated patients. The mechanism of fat graft survival is not clear, and the role of adiposederived stem cells and preadipocytes in fat survival remains to be determined.23,24 A correct technique, as described (fan shape, multitunnel, etc.) could avoid clinical oil cyst to form. Our five cases of cystoteatonecrosis appeared at the beginning of our experience. Therefore, a short training period is required for surgeons performing lipomodelling. They must be able to transfer important volumes of fat. The rate of fat resorption is approximately and subjectively, around 30%, in our experience. Therefore, the graft must represent at least 140% of the target volume. Besides, with only 70% of pure fat graft obtained after centrifugation, the volume harvested must be approximately 200% of the target volume. This implies harvesting many syringes of fat, which takes time. To keep the duration of the procedure to a minimum, it is advisable to perform the different steps simultaneously: the surgeon harvests the fat graft, then a surgical assistant prepares the graft for centrifugation and injection while the surgeon proceeds with further steps of the procedure, such as breast symmetrisation. In agreement with previous reports,25e28 it seems that the buttocks and the trochanter are the highest yield sites, with more than 80% fat graft. However, these results remain controversial.6,28 By contrast, harvesting fat from these sites implies that the patient be moved, which makes the procedure longer. Whenever possible, we therefore prefer to harvest fat graft from the abdomen.
1776 Major postoperative complications of lipomodelling are rare.22,23,29,30 As for liposuction, patients complain of pain and oedema. Only three (1.5%) major complications were reported in our series of 200 patients, with two infections at the graft site, which were easily controlled without compromising patient outcome or delaying recovery, and one case of pneumothorax requiring simple pleural drainage. The most serious drawback in our series was graft resorption. The rates of resorption and survival of the fat graft grafted to the patients are hardly predictable. We subjectively used rates around 30%, whereas other reports have used values between 30% and 80%.7 We are currently developing an evaluation system for the objective assessment of fat resorption after lipomodelling.31,32 This 3D surface scan system provides a 3D model of the patient. Comparing images taken before and after the procedure makes it possible to calculate volume differences and find out the rate of fat resorption. Preliminary results indicate rates between 30% and 40%, according to the patient, reduced by approximately 20% at the second lipomodelling session.
Future prospects Lipomodelling is a useful complement to autologous latissimus dorsi flap reconstruction of the breast. Using this procedure, it has been possible to extend the indications for flap reconstruction in our patients (95%) and also to improve the quality of breast reconstruction.14 Currently, we propose the latissimus dorsi flap to only those patients with enough fat on the dorsal thoracic area. Now, with the second stage of lipomodelling, we can complete the lack of volume during the second surgery and also propose this technique to more patients. The principle of latissimus dorsi flap reconstruction has been modified and the autologous tissue flap is now used as a recipient site for further fat cell injections. Preliminary radiological studies have demonstrated the absence of deleterious consequences for breast radiological follow-up, thus permitting to extend the indications for breast lipomodelling.13,33e35 We have tested the procedure in patients requiring only partial breast reconstruction, or in those with breast deformities (pectus excavatum and Poland’s syndrome), tuberous breasts, unilateral hypotrophy or asymmetric breasts.14,15,36,37 From the results obtained in these patients, it seems reasonable to extend the use of lipomodelling to women seeking cosmetic breast surgery.19 From the present study, we conclude that breast lipomodelling is safe and effective for patients undergoing breast reconstruction with autologous latissimus dorsi flap. Grafting fat graft to the reconstructed breast helps improve the very satisfactory results obtained with primary plastic surgery, which makes autologous flap reconstruction the treatment of choice for our patients. Using fat graft taken from the patient helps preserve the autologous nature of the procedure. Autologous latissimus dorsi flap placement can be seen as the first step of the reconstruction, which prepares the ground for further enhancement using lipomodelling. Lipomodelling is simple, safe and reproducible, which makes it available for use in many indications of breast
R. Sinna et al. reconstructive and plastic surgery. The technique thus appears as one of the most considerable advances in breast reconstructive surgery since the mid-1990s.
Conflict of interest None.
Funding source None.
References 1. Delay E, Jorquera F, Pasi P, et al. Autologous latissimus breast reconstruction in association with abdominal advancement flap: a new refinement in breast reconstruction. Ann. Plast. Surg 1999;42:67e75. 2. Delay E, Jorquera F, Lucas R, et al. Sensitivity of breasts reconstructed with the autologous latissimus dorsi flap. Plast. Reconstr. Surg 2000;106:302e9. 3. Delay E, Gounot N, Bouillot A, et al. Breast reconstruction with the autologous latissimus dorsi flap. Preliminary report of 60 consecutive reconstructions. Ann. Chir. Plast. Esthet 1997;42: 118e30. 4. Delay E, Delaporte T, Sinna R. Breast implant alternatives. Ann. Chir. Plast. Esthet 2005;50:652e72. 5. Coleman SR. Long-term survival of fat transplants: controlled demonstrations. Aesthetic Plast. Surg 1995;19:421e5. 6. Chajchir A. Fat injection: long-term follow-Up. Aesthetic Plast. Surg 1996;20:291e6. 7. Sinna R, Delay E, Garson S, et al. Scientific bases of fat transfer. Critical review of the literature. Ann. Chir. Plast. Esthet 2006;51:223e30. 8. Coleman SR. Facial recontouring with lipostructure. Clin Plast Surg 1997;24:347e67. 9. Missana MC, Laurent I, Barreau L, et al. Autologous fat transfer in reconstructive breast surgery: indications, technique and results. Eur J Surg Oncol 2007;33:685e90. 10. Spear SL, Wilson HB, Lockwood MD. Fat injection to correct contour deformities in the reconstructed breast. Plast Reconstr Surg 2005;116:1300e5. 11. Bircoll M. Cosmetic breast augmentation utilizing autologous fat and liposuction techniques. Plast. Reconstr. Surg 1987;79: 267e71. 12. ASPRS Ad-Hoc Committee on New Procedures. Report on autologous fat transplantation, September 30, 1987. 13. Pierrefeu-lagrange AC, Delay E, Guerin N, et al. Radiological evaluation of breasts reconstructed with lipomodelling. Ann. Chir. Plast. Esthet 2006;51:18e28. 14. Delay E. Lipomodelling of the reconstructed breast. In: Spear SL, editor. Surgery of the breast: principles and art. Philadelphia, Pa: Lippincott-Raven Publishers; 2006. 15. Delaporte T, Delay E, Toussoun G, et al. Breast volume reconstruction by lipomodeling technique: about 15 consecutive cases. Ann. Chir. Plast. Esthet 2009;54:303e16. 16. Clough KB, Louis-Sylvestre C, Fitoussi A, et al. Donor site sequelae after autologous breast reconstruction with an extended latissimus dorsi flap. Plast. Reconstr. Surg 2002;109: 1904e11. 17. Neuber F. Fetttransplantation. Chir kongr vehandl dtsch ges chir 1893;22:66. 18. Czerny V. Plasticher erzatz der Brustdruse durch ein lipom. Zentralb chir 1895;27:72.
Breast fat grafting (lipomodelling) after extended latissimus dorsi flap breast reconstruction 19. Zheng DN, Li QF, Lei H, et al. Autologous fat grafting to the breast for cosmetic enhancement: experience in 66 patients with long-term follow-up. J Plast Reconstr Aesthet Surg 2008; 61:792e8. 20. Guerrerosantos J, Gonzalez-Mendoza A, Masmela Y, et al. Long-term survival of free fat grafts in muscle: an experimental study in rats. Aesthetic Plast. Surg 1996;20:403e8. 21. Hoerl HW, Feller AM In: Shiffman MA, editor. Autologous fat volume retention: evaluation by magnetic resonance imaging, in autologous fat transplantation. New York: Marcel Dekker, Inc; 2001. p. 31e42. 22. Chajchir A, Benzaquen I. Fat-grafting injection for soft-tissue augmentation. Plast. Reconstr. Surg 1989;84:921e34. 23. Coleman SR. Structural fat grafting: more than a permanent filler. Plast Reconstr Surg 2006;118:108Se120S. 24. Rigotti G, Marchi A, Galie ´ M, et al. Clinical treatment of radiotherapy tissue damage by lipoaspirate transplant: a healing process mediated by adipose-derived adult stem cells. Plast Reconstr Surg 2007;119:1409e22. 25. Pinski KS, Roenigk Jr HH. Autologous fat transplantation. Longterm follow-up. J Dermatol. Surg. Oncol 1992;18:179e84. 26. Asken S. Autologous fat transplantation: micro and macro techniques. Am J Cosm Surg 1987;4:111. 27. Hudson DA, Lambert EV, Bloch CE. Site selection for fat autotransplantation: some observations. Aesthetic Plast. Surg 1990;14:195e7. 28. Rohrich RJ, Sorokin ES, Brown SA. In search of improved fat transfer viability: a quantitative analysis of the role of centrifugation and harvest site. Plast. Reconstr. Surg. 2004;113:391e5.
1777
29. Illouz YG. Present results of fat injection. Aesthetic Plast. Surg 1998;12:175e81. 30. De Pedroza LV. Fat transplantation to the buttocks and legs for aesthetic enhancement or correction of deformities: long-term results of large volumes of fat transplant. Dermatol. Surg 2000; 26:1145e9. 31. Garson S, Delay E, Sinna R, et al. 3D evaluation and breast plastic surgery: preliminary study. Ann. Chir. Plast. Esthet 2005;50:296e308. 32. Sinna R, Garson S, Taha F, et al. Evaluation of 3D numerisation with structured light projection in breast surgery. Ann. Chir. Plast. Esthet 2009;50:317e30. 33. Amar O, Bruant-Rodier C, Lehmann S, et al. Fat tissue transplant: restoration of the mammary volume after conservative treatment of breast cancers, clinical and radiological considerations. Ann Chir Plast Esthet 2008;53:169e77. 34. Carvajal J, Patin ˜o JH. Mammographic findings after breast augmentation with autologous fat injection. Aesthet Surg J 2008;28:153e62. 35. Gosset J, Guerin N, Toussoun G, et al. Radiological evaluation after lipomodelling for correction of breast conservative treatment sequelae. Delay E. Ann Chir Plast Esthet 2008 Apr; 53:178e89. 36. Pinsolle V, Chichery A, Grolleau JL, et al. Autologous fat injection in Poland’s syndrome. J Plast Reconstr Aesthet Surg 2008;61:784e91. 37. Delay E, Gosset J, Toussoun G, et al. Efficacy of lipomodelling for the management of sequelae of breast cancer conservative treatment. M. Ann Chir Plast Esthet 2007;53:153e68.