Cell-Assisted Lipotransfer (CAL) for Cosmetic Breast Augmentation -Supportive Use of Adipose-derived Stem/Stromal Cells- Kotaro Yoshimura, M.D.,1 Katsujiro Sato, M.D.,2 Noriyuki Aoi, M.D.,1 Masakazu Kurita, M.D.,3 Toshitsugu Hirohi, M.D.,4 and Kiyonori Harii, M.D.3

Introduction
Autologous fat transplantation is one of the promising treatments for facial rejuvenation and soft-tissue augmentation due to the lack of incisional scar and complications associated with foreign materials. However, certain problems remain, such as unpredictability and a low rate of graft survival due to partial necrosis. Many innovations have been reported in an effort to overcome these problems [1, 2, 4-6, 18] and reviewed previously [4, 14]. Based on these reports we tentatively concluded that we could harvest fat with a 2.5 mm cannula or 18-gauge needle at less than 700 mmHg vacuum and re-inject it with an 18-gauge needle without significant adipocyte damage [14].
Lipoinjection can be used for treating facial changes associated with aging, correcting various kinds of depressed deformities such as hemifacial microsomia and pectus excavatum., It has also been used in breast augmentation by a limited number of plastic surgeons [3], although the use of autologous fat for breast augmentation has been controversial due to the lack of consensus on whether it is safe and appropriate because of microcalcifications that may cause confusion in the evaluation of mammograms. Recently, autologous fat injection has been re-evaluated as a potential alternative to artificial implants for breast augmentation [3, 15, 16, 19]. This re-evaluation may reflect recent advances in autologous fat transfer and the radiological detection of breast cancer.
To overcome the problems associated with autologous fat transfer, we have used a novel strategy, called Cell-Assisted Lipotransfer (CAL) (Fig. 1). Tissue-specific progenitor cells in the adipose tissue were found to have the capacity to differentiate into various cell lineages [21]. Thus, the progenitors are now called “adipose-derived stem/stromal cells (ASCs)”, and expected to become a valuable tool in a wide range of cell-based therapies. The therapeutic concept of CAL was described in our previous report on pre-clinical studies [9]. We found that aspirated fat has approximately half the number of ASCs than excised whole fat. The two main reasons for this relative deficiency are 1) a major portion of the ASCs are located around large vessels and left in the donor site after liposuction [9], and 2) a part of ASCs are released into the fluid portion of liposuction aspirates [14]. The relative deficiency of ASCs may induce postoperative long-term atrophy of injected fat, as was partially confirmed in animal studies [8, 9, 11]. In the CAL strategy, autologous ASCs are used to enhance angiogenesis, improve the survival rate of grafts, and reduce post-operative atrophy. In CAL, half the volume of the aspirated fat is processed for isolation of the stromal vascular fraction (SVF) containing ASCs. During the isolation process, the other half of the aspirated fat is prepared for grafting. Freshly isolated SVF, which we characterized before [20], is attached to the aspirated fat with the fat acting as a living scaffold before transplantation. Finally, the SVF-supplemented fat is injected to the target sites. Thus, ASC-poor fat is converted to ASC-rich fat in the preparation process of the injectable material. In this study, we report on the preliminary results in patients who underwent CAL for cosmetic breast augmentation. This is the first report of clinical use of ASCs for cosmetic purposes.

Materials and Methods
Patients
From 2003 to 2007, we have performed CAL in 70 patients; in the breast in 60 patients (including 8 patients for breast reconstruction after mastectomy), the face in 12 patients, and the hip in 1 patient (CAL was performed at 2 sites in 3 patients). Informed consent was obtained from all patients. The study protocol conformed to the guidelines of the 1975 Declaration of Helsinki and was approved by individual institutional review boards.
In this study, 40 patients who had healthy thoraxes and breasts underwent CAL for purely cosmetic breast augmentation (i.e. breast reconstruction for inborn anomaly or after mastectomy was not included). Nineteen of these 40 patients were followed for more than 6 months (at the time of this report) and the maximum follow-up period has been 42 months. All of the patients were Japanese females with a body-mass index of 19.1 ± 1.9 (mean ± standard deviation) and the patient’s ages varied from 20 to 62 years (35.8 ± 9.1). The mean volume of injected fat was 268.1 ± 47.6 ml on the left side, and 277.3 ± 39.1 ml. Demographic and surgical data on the patients are summarized in Table 1.

Surgical Techniques
Before the procedure began, the liposuction site was infiltrated with saline solution with diluted epinephrine (0.001%). With the patient under general anesthesia, adipose tissue was suctioned using a cannula with 2.5-mm inner diameter and a conventional liposuction machine. . About a half of the collected liposuction aspirate was used for isolation of the SVF. The SVF was isolated from both the adipose portion and the fluid portion of liposuction aspirates as described previously [20] and the cell processing procedure took about 90 min. During the processing period, the other half of lipoaspirates was harvested as graft material.
The adipose portion of liposuction aspirates was either 1) washed several times and placed in an upright position for obtaining clear separation from fluids and oil (Group A and B), or 2) centrifuged at 700 g for 3 min without washing (Group C), and put into a metal jar (500 ml) which was placed in water with crushed ice. In Groups A and C, the fresh SVF isolated from both the adipose and fluid portion was added to the graft material and, after gentle mixing and waiting for 10-15 min for cell adherence to the aspirated fat, the cell-supplemented fat was then put into an injection syringe. In Group B, the freshly isolated SVF was resuspended in 60ml of saline, and diffusely injected into the whole breast mounds (30 ml for each breast) separately, just after conventional lipoinjection. The patient numbers of Groups A, B, and C were 6, 2, and 32, respectively.
For the injection syringe, a 10 cc LeVeen? inflator (Boston Scientific Corp., MA) or our original syringe (20 ml) was used because they are screw-type syringes (with a threaded plunger) and threaded connections that fit both the connecting tube and the needle, to allow for precise control during injection (Fig. 2A). To reduce the time of the procedure, two syringes were used; while one syringe was being used for an injection, the other was filled with the graft material in preparation for the next injection. An 18-gauge needle (150 mm long) was used for lipoinjection and inserted subcutaneously at one of 4 points indicated in Fig. 2B. The operator took care to insert and place the needle horizontally (parallel to the body), in order to avoid damaging the pleura and causing a pneumothorax. The needle was inserted in several layers and directions, and was continuously and gradually retracted as the plunger was advanced. This technique was used to obtain a diffuse distribution of the graft material (Figs. 2B, and 3). The grafts were placed into the fatty layers on, around, and under the mammary glands, and also into the pectoralis muscles.

Results
The transplantation of adipose tissue was successfully performed in all cases, and the time of the injection process ranged from 35 to 60 min for both breasts. Subcutaneous bleeding was occasionally seen on some parts of the breasts, and resolved in one to two weeks.
Transplanted adipose tissue was gradually absorbed during the first 2 postoperative months (especially during the first month), and the breast volume showed a minimal change thereafter, though skin tension sometimes became looser after 2 months. Photograph of three representative surgical sites are shown in Figures 4 to 9. Breast circumference difference (= chest circumference at the nipple ? chest circumference at the infra-mammary fold) increased in all cases, by 4 to 8 cm at 6 months, which corresponds to 2 to 3 cups sizes of brassiere. The increase in the circumference seems to correspond to 100-200 ml increase in the volume of each breast mound, which was partially confirmed by our preliminary evaluation using a 3-dimesional quantitative measurement system. Compared to breast augmentation with implants of the same size, augmentation with CAL showed a lower height but more natural contour of breasts. All cases but one (see below) showed natural softness of the breasts without any palpable nodules at 6 months, and all patients were satisfied with the resulting texture, softness, contour and absence of foreign materials despite the limited size increase possible with autologous tissue. Cyst formation (< 12 mm) was detected by MRI in 2 patients, and micro-calcification was detected by mammogram in 2 patients at 24 months. In one of 2 patients in group B, fibrous breast tissue and fibrosis on the sternum were observed by CT scan at 6 months, and the breasts were found to be harder than other cases.

Discussion
A number of modifications of lipoinjection techniques have been attempted in order to improve the survival rate of injected fat. From these, it is well accepted that adipose tissue should be placed as small aliquots [3], preferably within an area 3 mm in diameter [1]. Since it takes a long time to perform ideally diffuse distribution of suctioned fat [3], we have used a disposable syringe with a threaded plunger and connections, a very long needle (150 mm), and an assistant to rotate the plunger, leading to only 35-60 min for injection in both breasts. These devices are critical to performing large-volume　lipoinjection safely and precisely in a short time.
In addition, harvesting, preserving, and refining graft materials are also important, as repeatedly indicated in the literature. We used a relatively large-sized suction cannula, centrifuged the aspirated fat in some cases, and kept it cooled until transplantation. In this study, clinical results (increase in breast circumference) appeared to be superior in Group C using centrifuged fat to Group A using non-centrifuged fat, though quantitative measurement and statistical comparison were not done. In a previous study, we found that centrifugation of aspirated fat is substantially influential because centrifugation at 1,200 g decreases the fat volume by 30%, damages 12% of the adipocytes and 0% of the ASCs, which leads to the concentration of cell numbers per volume of adipocytes and ASCs by 25% and 43%, respectively [7]. In addition, centrifugation may be especially beneficial in our treatment, because water content in the graft material may disturb the adherence of ASCs to the adipose tissue and interfere with differentiation into expected lineages. ASCs floating in a solution, which is a non-physiological environment, may migrate over distances, penetrate into the lymphatic flow, and differentiate unexpectedly. We believe that such migration and altered cell-differentiation caused the development of fibrotic tissue on the sternum of one patient in Group B. Thus, we conclude that centrifuged fat combined with ASCs as cell pellets (i.e. Group C) was best among the three methods used in this study.
Although small cystic formation and micro-calcification were detected in some cases, the micro-calcification was easily distinguished from that associated with breast cancer and the overall cosmetic results were generally satisfactory and encouraging. Almost all patients were satisfied with their enlarged and soft breasts with a natural contour. CT scans and MRI showed that transplanted fat tissue survived and formed a significant thickness of the fatty layer not only subcutaneously on and around the mammary glands but also between the mammary glands and the pectoralis muscles. Breast volume stabilized 2 to 3 months after transplantation. Maximum breast augmentation with this technique varied among patients and appeared to be 100-200 ml. While these volumes may be smaller than those achieved with large artificial implants, a definite advantage is that patients do not have to be concerned about postoperative complications induced by artificial implants such as rupture, infection, capsular contracture, unnatural contour, hardness, neurological symptoms and immune response. Compared to our dozens of patients who underwent conventional autologous lipoinjection to the breasts, augmentation effects were apparently higher in CAL; a 2-3 cm increase in breast circumference was common in the conventional procedure, but 4-8 cm increase was seen in this trial of CAL, though the augmentation effect varied among patients. The measurement system we recently devised may help to quantify the difference in augmented volume in the future.
It has been revealed that adipose tissue contains not only adipogenic progenitor cells but multipotent stem cells which can differentiate into fat, bone, cartilage, and types of tissue [21, 22]. Suctioned fat appears to lose a significant number of these precursors during liposuction and the preparation processes as compared to non-suctioned adipose tissue [9]. This relative deficiency of precursors may contribute to the low survival rate and long-term atrophy of transplanted lipoaspirates. In CAL, the deficit of ASCs was compensated for by supplementing ASCs. In order to maximize the biological function and avoid unexpected behavior of ASCs, it seems important to ensure adherence of supplemented ASCs to adipocytes or connective tissue. There are four possible roles for ASCs in this novel treatment, which were partly confirmed in pre-clinical studies [8, 9, 11]. First, ASCs can differentiate into adipocytes and contribute to regeneration of adipose tissue. Second, ASCs can differentiate into endothelial cells and also probably into vascular mural cells [8, 10,12], resulting in the promotion of angiogenesis and graft survival. Third, ASCs are known to release angiogenic growth factors in response to hypoxia and other conditions [13], and these factors influence surrounding host tissue. The last role, which may be the most influential, is that ASCs survive as original ASCs [9]. In the adipose, ASCs reside between adipocytes or in the extracellular matrix, especially around vessels, and contribute to the turnover of adipose tissue, which is known to be very slow (2 years or more) [17]. However, adipose grafts probably turn over during the first 2 to 3 months after transplantation, because they experience temporary ischemia followed by reperfusion injury. This turnover, the replacement process of the adipose tissue, will be conducted by tissue-specific progenitor cells, which are ASCs. The relative deficiency of ASCs in aspirated fat may affect the replacement process and lead to post-operative atrophy of grafted fat, which is known to commonly occur during the first 6 months after lipoinjection.
The freshly isolated SVF used in CAL contains not only ASCs but also vascular endothelial cells, pericytes, blood cells (WBCs and RBCs), and other cells as previously described [20]. ASCs may interact with other cells after transplantation, such as vascular endothelial cells, and supplementation with the SVF may be superior to ASCs alone in this treatment. However, further studies are needed to elucidate the synergistic effects of ASCs with other cells contained in the graft.
In this preliminary study, satisfactory clinical results were generally achieved without any major complications. Thus we can conclude that CAL is safe enough to continue the study though controlled studies and the accumulation of long-term results are needed to elucidate the overall safety and efficacy of the treatment. A variety of new innovations including stem cell technology may be developed and contribute to the improvement of autologous tissue transplantation and regeneration. Further improvements of the technique may cause autologous tissue transfer to become the first choice for breast augmentation in the future.

References
1. Carpaneda CA, Ribeiro MT (1994) Percentage of graft viability versus injected volume in adipose autotransplants. Aesthetic Plast Surg 18: 17-19
2. Coleman SR (2001) Structural fat grafts: the ideal filler? Clin Plast Surg 28: 111-119
3. Coleman SR, Saboeiro AP (2007) Fat grafting to the breast revisited: safety and efficacy. Plast Reconstr Surg 119: 775-785
4. Ersek RA, Chang P, Salisbury MA (1998) Lipo layering of autologous fat: an improved technique with promising results. Plast Reconstr Surg 101: 820-826
5. Fagrell D, Enestrom S, Berggren A, Kniola B (1996) Fat cylinder transplantation: an experimental comparative study of three different kinds of fat transplants. Plast Reconstr Surg 98: 90-96
6. Har-Shai Y, Lindenbaum ES, Gamliel-Lazarovich A, Beach D, Hirshowitz B (1999) An integrated approach for increasing the survival of autologous fat grafts in the treatment of contour defects. Plast Reconstr Surg 104: 945-954
7. Kurita M, Matsumoto D, Shigeura T, Sato K, Gonda K, Harii K, Yoshimura K. Influences of centrifugation on cells and tissues in liposuction aspirates: optimized centrifugation for lipotransfer and cell isolation. Plast Reconstr Surg, in press.
8. Masuda T, Furue M, Matsuda T (2004) Novel strategy for soft tissue augmentation based on transplantation of fragmented omentum and preadipocytes. Tissue Eng 10: 1672-1683
9. Matsumoto D, Sato K, Gonda K, Takaki Y, Shigeura T, Sato T, Aiba-Kojima E, Iizuka F, Inoue K, Suga H, Yoshimura K (2006) Cell-assisted lipotransfer: supportive use of human adipose-derived cells for soft tissue augmentation with lipoinjection. Tissue Eng 12: 3375-3382
10. Miranville A, Heeschen C, Sengenes C, Curat CA, Busse R, Bouloumie A (2004) Improvement of postnatal neovascularization by human adipose tissue-derived stem cells. Circulation 110: 349-355
11. Moseley TA, Zhu M, Hedrick MH (2006) Adipose-derived stem and progenitor cells as fillers in plastic and reconstructive surgery. Plast Reconstr Surg 118(3 Suppl):121S-128S
12. Planat-Benard V, Silvestre JS, Cousin B, Andre M, Nibbelink M, Tamarat R, Clergue M, Manneville C, Saillan-Barreau C, Duriez M, Tedgui A, Levy B, Penicaud L, Casteilla L (2004) Plasticity of human adipose lineage cells toward endothelial cells: physiological and therapeutic perspectives. Circulation 109: 656-663
13. Rehman J, Traktuev D, Li J, Merfeld-Clauss S, Temm-Grove CJ, Bovenkerk JE, Pell CL, Johnstone BH, Considine RV, March KL (2004) Secretion of angiogenic and antiapoptotic factors by human adipose stromal cells. Circulation 109:1292-1298
14. Shiffman MA, Mirrafati S (2001) Fat transfer techniques: the effect of harvest and transfer methods on adipocyte viability and review of the literature. Dermatol Surg 27: 819-826
15. Spear SL, Wilson HB, Lockwood MD (2005) Fat injection to correct contour deformities in the reconstructed breast. Plast Reconstr Surg 116: 1300-???
16. Spear SL, Newman MK (2007) Discussion to “Fat grafting to the breast revisited: safety and efficacy”, Plast Reconstr Surg 119: 786-787
17. Strawford A, Antelo F, Christiansen M, Hellerstein MK (2004) Adipose tissue triglyceride turnover, de novo lipogenesis, and cell proliferation in humans measured with 2H2O. Am J Physiol Endocrinol Metab 286, E577-E588
18. Ullmann Y, Hyams M, Ramon Y, Peled IJ, Leiderbaum ES (1998) Enhancing the survival of aspirated human fat injected into nude mice. Plast Reconstr Surg 101: 1940-1944
19. Yoshimura K, Matsumoto D, Gonda K (2005) A clinical trial of soft tissue augmentation by lipoinjection with adipose-derived stromal cells (ASCs). Proceedings of the 3rd annual meeting of International Fat Applied Technology Society (IFATS), pp.9-10, Charlotteville, Virginia.
20. Yoshimura K, Shigeura T, Matsumoto D, Sato T, Takaki Y, Aiba-Kojima E, Sato K, Inoue K, Nagase T, Koshima I, Gonda K (2006) Characterization of freshly isolated and cultured cells derived from the fatty and fluid portions of liposuction aspirates. J Cell Physiol 208: 64-76
21. Zuk PA, Zhu M, Ashjian P, De Ugarte DA, Huang JI, Mizuno H, Alfonso ZC, Fraser JK, Benhaim P, Hedrick MH (2002) Human adipose tissue is a source of multipotent stem cells. Mol Biol Cell 13: 4279-4295
22. Zuk PA, Zhu M, Mizuno H, Huang J, Futrell JW, Katz AJ, Benhaim P, Lorenz HP, Hedrick MH (2001) Multilineage cells from human adipose tissue: implications for cell-based therapies. Tissue Eng 7: 211-228

Legends

Fig. 1. Scheme of Cell-Assisted Lipotransfer. Relatively ASC-poor aspirated fat is converted to ASC-rich fat by supplementing ASCs isolated from the other half of the aspirated fat. The ASCs are attached to the aspirated fat, which is used as a scaffold in this strategy.

Fig. 2. Schematic instruction of the injection method. (A) A small amount of fat tissue is injected as small aliquots or a thin string with a long needle on a syringe with a threaded plunger while the needle is continuously withdrawn. (B) The needle is inserted from either one of two points on the areola margin or one of two points at the infra-mammary fold in variable directions and planes to achieve a diffuse distribution.

Fig. 3. A clinical view of injection. The injection needle is rigidly manipulated by an operator, while an assistant rotates the plunger according to the operator’s instruction. A high-pressure injection can be performed with a disposable syringe with a threaded plunger. A 150 mm-long 18-gauge needle is connected to the syringe with a connecting tube threaded at both ends.

Fig. 4. Clinical views of a patient in Group A (Patient #1); Preoperative views (top) and postoperative views at 24 months (bottom). A twenty-two-year-old woman underwent breast augmentation with CAL (290 ml in each breast) with satisfactory results at 24 months. Her breast circumference increased by 5.0 cm. Augmented breast mounds remained soft and natural appearing without injection scars or subcutaneous indurations.


Fig. 5. Radiological views of Patient #1’s chest. (A) A preoperative CT image in the horizontal plane of the nipples. (B) A horizontal image 12 months after surgery. Note that the adipose tissue is augmented both subcutaneously and under the mammary glands. (C) Mammograms at 12 months show no calcification or other abnormal signs.

Fig. 6. Clinical views of a patient in Group C (Patient #2); Preoperative views (top) and postoperative views at 12 months (bottom). A thirty-two-year-old woman underwent breast augmentation with CAL (280 ml in each breast). Her breast circumference difference increased from 9.0 cm (baseline) to 14.5 cm (at 12 months). The breast mounds are soft and natural appearing with no visible injection scars.

Fig. 7. Radiological views of Patient #2’s chest. (A) A preoperative CT image in the horizontal plane at the level of the nipples. (B and C) Horizontal images by MRI (B, T1-image; C; T2-image) 12 months after surgery. The adipose tissue is augmented around and under the mammary glands. A small (< 10 mm) cyst appears in the fatty layer under the right mammary gland. (D) Mammograms at 12 months show no abnormal signs such as calcifications.

Fig. 8. Clinical views of a patient in Group C (Patient #3); Preoperative views (top) and postoperative views at 24 months (bottom). A thirty-old woman underwent breast augmentation with CAL (310 ml in each breast). Her breasts were dramatically augmented with an increase in breast circumference difference by 8.0 cm at 24 months. The breast mounds were soft with no subcutaneous indurations. An original infra-mammary fold on the left breast is slightly visible, but injection scars are not visible.

Fig. 9. Radiological views of Patient #3. (A) A preoperative CT image in the horizontal plane at the level of the nipples. Only a very thin fatty layer is observed around the mammary glands. (B) A horizontal MRI image (T1 weighted) 24 months after surgery. Transplanted adipose tissues survived and formed thick layers around and under the mammary glands. (C) Mammograms 24 months after surgery show no abnormal signs.