Adipose tissue has been considered an organ of energy storage, the largest endocrine organ, a soft tissue filler (augmentation by micro-fat grafting), and a cosmetically unnecessary tissue discarded by liposuction. Adipose tissue physiologically turns over very slowly. Adipocytes have a life span of 10 years;¹ several thousands of adipocytes die and are replaced with new adipocytes every second in our body. Adipose tissue has many cells other than adipocytes, though adipocytes constitute more than 90% of adipose tissue volume. Through collagenase digestion, cellular components other than adipocytes are extracted as a cell pellet, which is called stromal vascular fraction (SVF). SVF contains adipose stromal cells (ASC), vascular endothelial cells (VEC), pericytes, adipose-resident macrophages, etc. (Fig. 48.1). It was found that the adipose stromal cell population contains not only monopotent progenitor cells responsible for the tissue turnover, but also multipotent mesenchymal stem cells,² which are now called adipose (-derived) stem cells. These are regarded as a potent tool for cell-based therapies, comparable to bone marrow-derived mesenchymal stem cells, because they can be obtained in a large amount through a less invasive approach. For adipose tissue engineering/regeneration, it is a critical point to understand the physiological roles and functions of ASC; this population, working together with VEC and stem/progenitor cells recruited from bone marrow, is a main player for adipose tissue engineering, which means adipogenesis and angiogenesis.

FIG. 48.1 From liposuction to isolation of adipose stem/progenitor cells.

Stromal vascular fraction (SVF) can be obtained through collagenase digestion of the lipoaspirates. SVF contains many blood-derived cells (CD45⁺) and adipose-derived cells (CD45⁻); the latter is further divided into CD31⁺/CD34⁺ endothelial cells (ECs), CD31⁻/CD34⁺ adipose progenitor cells (ASCs) and CD31⁻/CD34⁻ other cells.

Autologous fat transplantation is one of the promising cosmetic treatments for facial rejuvenation and soft-tissue augmentation due to minimal incisional scars and lack of complications associated with foreign materials. However, certain problems remain, such as unpredictability and a low rate of graft survival due to partial necrosis. Implantation of prostheses has been a golden standard for breast augmentation, but complications such as capsular contracture remain to be resolved. Recently, autologous fat injection has been re-evaluated as a potential alternative to artificial implants for breast augmentation.³ This re-evaluation may reflect recent advances in autologous fat transfer and the radiological detection of breast cancer.

In this chapter, biological principles and surgical techniques of fat injection for breast augmentation are discussed as well as our novel approach of autologous fat grafting called cell-assisted lipotransfer (CAL),^4,⁵ which is concurrent transplantation of aspirated fat tissue and adipose progenitor cells (Fig. 48.2).

FIG. 48.2 Scheme of cell-assisted lipotransfer (CAL).

Relatively adipose-derived stromal/stem cells (ASC)-poor aspirated fat tissue is converted to ASC-enriched tissue by supplementation with ASCs isolated from a part of the lipoaspirates. The ASCs are attached to the aspirated fat tissue, which is used as a scaffold in this strategy. SVF, stromal vascular fraction.

Preoperative Preparation

There seem to be several patient factors that may affect the clinical result of lipoinjection; skin redundancy of the breasts, age, BMI, personal quality or character of fat tissue, adhesive scars, breast implant and its capsule, systemic disease such as autoimmune disease, oral corticosteroid, etc.^4,⁶ Good candidates are those who have sufficient fat at donor sites and the redundant breast skin with healthy vascularity and without scars. Lipoinjection can be performed in any patients from teens to 70s, but patients with low BMI (less than 18) are not good candidates due to difficulty in harvesting a sufficient volume of fat tissue without severe morbidity. Patients who want a large-volume (250–400 ml) augmentation, are not good candidates because augmentation volume achieved by a single session of lipoinjection is limited (100–200 ml).

No particular preparation is generally needed preoperatively. However, for breast reconstruction patients with adhesive scars, an external negative pressure device (Brava®, Brava LLC, Miami FL, USA) possibly improve clinical outcomes. For young patients with small breasts and tight skin, injection volume is limited due to the tight skin envelope. In these cases, a secondary lipoinjection (replacement with fat tissue after temporary breast implant) would be recommended. This allows for her breast skin to be stretched by the implants.

For preoperative examination, physical measurements (maximum and bottom breast circumferences, etc.), mammogram, MRI, echogram, photograph, and videograph are performed. We also adopted a 3-dimensional measurement system that enables a volumetric evaluation of the breast mound in a standing position (Fig. 48.3). An echogram is easy to perform at every patient’s visit and sensitive enough to detect small cyst formation. Long follow up with an annual mammogram up to 5 years is recommended to detect abnormal signs such as calcifications.

FIG. 48.3 Three-dimensional system for measuring breast volume.

Two cameras take photos with striped light shed on the breast. A software program calculates the volume over the standard plane. Using this system, breast volume can be measured while the patient is in a standing position.

Surgical Technique

Breast Augmentation

Donor sites are usually the thighs alone or the thighs and the abdomen or flanks, decided according to the patient’s preference and BMI. After the liposuction site is infiltrated with saline solution with epinephrine (0.001%) under general anesthesia, adipose tissue is suctioned using a cannula with 2–3 mm inner diameter and a conventional liposuction machine. We use screw-type syringes (with a threaded plunger) to allow for precise control and avoid a flushing injection (Fig. 48.4). To reduce the time of the procedure, two syringes are used; while one syringe is being used for an injection, the other is filled with the graft material in preparation for the next injection. A 16- or 18-gauge needle (150 mm long) is used for lipoinjection and inserted subcutaneously from the inframammary fold or areolar margin (Fig. 48.5A). The operator takes care to insert and place the needle horizontally (parallel to the body), in order to avoid causing a pneumothorax. The needle is inserted in several layers and directions, and is continuously and gradually retracted as the plunger is advanced (Fig. 48.5B). This technique is used to obtain a diffuse distribution of the graft material. The grafts are placed into the fatty layers on, around, and under the mammary glands (but not into the mammary glands), and also into the pectoralis muscles. After training, it is not hard for an operator to recognize the mammary gland or pectoralis fascia as a harder tissue than the fat or muscle tissue. Injection is discontinued when the skin tension becomes tense; injection volume is usually 200–300 ml for each breast in Asian patients.

FIG. 48.4 Injection devices.

A high-pressure injection can be performed with a disposable syringe with a threaded plunger. A 150-mm long 16- or 18-gauge needle is connected to the syringe with a connecting tube threaded at both ends. The injection needle is rigidly manipulated by an operator, while an assistant rotates the plunger according to the operator’s instruction. (A) a 10 cm³ LeVeen™ inflator (Boston Scientific Corp., MA); (B) a 20 cm³ screw injector original syringe (Medical U&A, Japan).

FIG. 48.5 Schematic instruction of the injection method for breast augmentation.

(A) The needle is inserted from either one of two points on the areola margin or one of two points at the inframammary fold in various directions and planes to achieve a diffuse distribution. (B) A small amount of fat tissue is injected as thin strings (1 ml fat corresponds to a 10 cm long string) with a long needle while the needle is continuously withdrawn.

Breast Implant Replacement

For patients with implants, lipoinjection can be performed simultaneously with implant removal.⁷ Breast implants are removed through a periareolar incision, which is placed at a caudal third of the areolar margin. The lipoinjection is begun at the deepest layer (under the implant capsule when the implant was subglandular) and completed with the injection into the most superficial subcutaneous layer. During the injection, the operator can insert fingers into the implant capsule and recognize the location of the injection needle (Fig. 48.6). Injection into the mammary glands or into the capsular cavity is not performed. Finally, the capsular cavity is irrigated with saline to remove any fat tissue injected and the periareolar incision is closed.

FIG. 48.6 Schematic illustration of the injection method for implant replacement.

(A) Using a long needle attached to a syringe with a threaded plunger, the fat tissue is injected into layers other than the mammary gland and capsular cavity. While injecting, the fingers are inserted through a periareolar skin incision into the implant capsule to determine the location of the needle tip. (B) The needle is inserted either through a periareolar incision or at one of several points on the infra-mammary fold. The needle is aimed in various directions and into different planes to achieve a diffuse distribution.

Preparation Procedures of Graft Materials

Conventional Lipoinjection

Lipoaspirates are harvested by suction-assisted liposuction and centrifuged (1200 g, 3 min), and put into a metal jar (1000 ml), which is placed in water with crushed ice. As the centrifugation reduces the adipose volume by 25–35%, the volume reduction should be taken into account in tissue harvesting (Fig. 48.7).

FIG. 48.7 Effects of centrifugation on aspirated adipose tissue.

The adipose portion was concentrated into 71% of the original volume after centrifugation for 3 minutes at 1200 g and the volume of the fluid and oil portions was significantly increased. However, the number of adipose-derived stromal cells (ASC) contained in the adipose portion was not significantly changed by the centrifugation. Thus, centrifugation at 1200 g led to condensation of cell numbers per volume of adipocytes and ASC by 25% and 43%, respectively, and improved the ASC : adipocyte ratio by 14%.

Cell-Assisted Lipotransfer (CAL)

Aspirated fat tissue has a significantly lower progenitor/mature-cell ratio,⁵ and this low ASC/adipocyte ratio may be one of reasons for long-term atrophy of transplanted adipose tissue as demonstrated before^4,^5,⁸ (Fig. 48.8). Enrichment of adipose progenitor cells by supplementation of SVF improves/normalizes progenitor/adipocyte ratio and is also suggested to augment final volume retention. In this strategy, progenitor-poor aspirated fat tissue is converted to progenitor-enriched (but not more than normal) fat tissue.

FIG. 48.8 Comparison of human aspirated and excised (intact) fat tissue obtained from a single site of a single patient.

Macroscopic views (A, B), schematic views (C, D), electromicroscopic (F, G, H and I; red scale bars = 200 µm, white scale bars = 40 µm), and whole mount staining images (E and J; scale bars = 100 µm). The basic structure of adipose tissue was preserved in the aspirated fat tissue, but some adipocytes and capillaries were disrupted. Vascular vessels, especially those of large size, were notably less in aspirated fat tissue compared to the excised fat tissue. ASC yield from aspirated fat tissue was considerably less (about a half) than that from excised fat tissue, suggesting that aspirated adipose tissue is relatively progenitor-poor fat tissue.

Full-CAL

In full-CAL, SVF is isolated from additionally harvested lipoaspirates (Fig. 48.9). The SVF is isolated as described below. During the processing period, liposuction is continued and lipoaspirates are centrifuged (700 g, 3 min) for preparation of graft materials. The freshly isolated SVF was added to the centrifuged fat tissue, followed by gentle mixing and a 5–10 min incubation to achieve appropriate cell adhesion to the centrifuged fat tissue (Fig. 48.10).

FIG. 48.9 Concept of cell-assisted lipotransfer (CAL).

The adipose portion (yellow) of liposuction aspirates is centrifuged and used as injection material in conventional lipoinjection. In CAL, relatively progenitor-poor aspirated fat tissue is converted to progenitor-enriched fat tissue by supplementation with the stromal vascular fraction (SVF). The SVF can be obtained from adipose and fluid portions of liposuction aspirates. In mini-CAL, which is used in thin patients, only the fluid portion (pink) of the liposuction aspirate is used for the isolation of the SVF, while another volume of liposuction aspirate is additionally harvested for SVF isolation in full-CAL, which is performed in fatty patients.

FIG. 48.10 Time course for CAL surgery.

The operation begins with the harvesting of fat tissue for the isolation of SVF cells. While the cell isolation process is performed in a cell-processing room, liposuction and centrifugation for graft material are performed in the operating room. After adding isolated SVF to the centrifuged fat tissue, the mixture is injected into the breast. The entire procedure takes approximately 3.5–4 hours to complete.

Mini-CAL

The difference of mini-CAL from full-CAL is that additional lipoaspirate is not harvested and only the fluid portion of liposuction aspirates is used for isolation of SVF (Fig. 48.9). This method is used when patients are thin and additional adipose tissue cannot be harvested.

SVF Isolation Procedure

Processed lipoaspirate cells (PLA) cells and liposuction aspirate fluid (LAF) cells are both so-called SVF; they are separated from the fatty and fluid portions of liposuction aspirates, respectively.⁹ PLA or both cells are used in full-CAL, while only LAF cells are used in mini-CAL (Fig. 48.9). For PLA cells, suctioned fat tissue is digested with 0.075% collagenase in phosphate-buffered saline (PBS) for 30 min on a shaker at 37°C. After centrifugation (800 g, 10 min), mature adipocytes and connective tissues (floating tissue) are discarded. The cell pellets are resuspended in Dulbecco’s modified Eagle’s medium (DMEM) and passed through a 100 mm mesh filter. To eliminate remaining collagenase, the cell pellets are repeatedly washed with resuspension in DMEM and following centrifugation at least three times. The process takes about 80 min (Fig. 48.10). For LAF cells, the suctioned fluid is centrifuged (400 g, 10 min) and the pellets are resuspended in distilled water (for 30 s) for erythrocyte lysis, followed by osmic normalization by adding 10% volume of 10× PBS (or 9% NaCl solution). After centrifugation (and filtration), cell pellets are obtained as SVF; the process takes about 20 min. For both PLA and LAF cells, cell counting for erythrocytes and nucleated cells is performed using a blood test cell counter; this takes 1 min.

Representative cases are shown in Figs 48.11–48.13.

FIG. 48.11 Case 1 (breast augmentation).

Preoperative views (A and B) and postoperative (C and D) views at 24 months. A 30-year-old woman underwent breast augmentation with CAL (310 ml in each breast). Her breasts were augmented dramatically with an 8.0 cm increase in breast circumference at 24 months. The breast mounds were soft with no subcutaneous indurations. An original inframammary fold on the left breast is slightly visible, but injection scars are not visible.

FIG. 48.12 Case 2 (breast augmentation).

Preoperative view (A) and postoperative (B) view at 12 months. A 36-year-old woman whose body mass index was 17.3 underwent breast augmentation with CAL (245 ml in each breast). The breast mounds were soft with no subcutaneous indurations or visible scars at 12 months.

FIG. 48.13 Case 3 (breast implant replacement).

A 33-year-old woman who had 210 ml saline implants underwent implant removal and simultaneous CAL (260 ml in each breast). The preoperative view showed capsular contractures and upward displacement of the left implant (A, C, E). At 12 months the breasts were symmetric and had a natural appearance (B, D, F). MRI at 12 months revealed that transplanted adipose tissues had survived and formed thick layers around and under the mammary gland. Mammograms showed no calcifications or other abnormal signs in either breast at 12 months. Augmented breast mounds maintained a sufficient breast volume even after implant removal and were naturally soft without any subcutaneous indurations.

Optimizing Outcomes

Understanding of Cellular Events in Fat Grafting

The grafted nonvascularized adipose tissue is placed under ischemia (hypoxia with low nutrition) and is temporarily nourished only by plasmatic diffusion from the surrounding host tissue for a few days until direct capillary supply is formed.⁹ In response to injury and cell death, many growth factors and proteinases are released from damaged tissue and activated platelets¹⁰ (Fig. 48.14). Inflammatory cells are infiltrated and inflammatory cytokines such as interleukins are secreted. During the repair process, adipocytes, known to be very sensitive to hypoxia, can die within 24 hours and subsequently endothelial cells start to die when the ischemia is prolonged. ASC are likely to be as resistant to ischemia as bone marrow-derived mesenchymal stem cells,¹¹ which can be functional up to 72 hours under severe ischemia (Fig. 48.15).

FIG. 48.14 Cellular and molecular events after adipose tissue grafting.

Adipose grafting induces injury in the recipient tissue; bleeding from the host tissue activates platelets, which release various soluble factors activating sleeping stem/progenitor cells. At the same time, transplanted adipose tissue is temporarily placed under severe ischemia; basic fibroblast growth factor (bFGF) and other factors are released from injured tissue or dying cells and stimulates adipose-derived progenitor cells to release hepatocyte growth factor (HGF), which promotes angiogenesis and inhibits fibrogenesis. Tissue injury further induces inflammatory cell infiltration and release of inflammatory cytokines. Most of the differentiated cells in the graft die, but graft-resident stem/progenitor cells are activated and bone marrow-derived stem/progenitor cells are recruited. The dead cells are partly replaced with next generation cells after successful revascularization. ECM, extracellular matrix; PDGF: platelet derived growth factor; EGF, epidermal growth factor; TGF-β, transforming growth factor-β; SDF-1, stromal derived factor-1; S1P, sphingosine-1-phosphate; IL, interleukin.

FIG. 48.15 Cellular events in ischemic adipose tissue.

Severe ischemia induces adipocyte death. If it is prolonged, vascular endothelial cells (VECs) also start to die. In contrast, adipose stem/progenitor cells can stay alive up to 72 hours even under severe conditions; during the first 72 hours, they are even activated and try to repair the damaged tissue by their regenerative efforts such as proliferation, migration and differentiation, collaborating with bone marrow-derived infiltrated cells.

Our recent studies showed that most adipocytes are likely to die within a few days (Fig. 48.16) and some of them are replaced with next generation adipocytes by 12 weeks.¹¹ From the surface of grafted fat tissue, adipocyte remodeling starts and progresses slowly toward the center and the rate of the adipocyte regeneration determines the final volume retention of fat grafting. In the case that stem/progenitor cells die due to prolonged severe conditions, the area is not properly replaced with new adipocytes and rather filled with fibrosis and calcification (Fig. 48.17). Oil cysts are being absorbed even after 3 months; the time course of the absorption depends on the size of lipid droplets, but takes weeks or months. Thus, nonvascularized adipose grafting can be called adipose tissue regeneration rather than adipose tissue survival, and we believe that this is seen in other nonvascularized tissue grafting such as skin or composite tissue grafting (Fig. 48.17).

FIG. 48.16 Histology of grafted adipose tissue at 4 weeks.

A section of adipose graft at 4 weeks stained with perilipin (alive adipocytes: green) and Hoechst (nuclei: blue) indicated that many adipocytes are dead (perilipin negative), which means they are lipid droplets (some of the dead adipocytes are indicated by asterisks). On the other hand, small new adipocytes (perilipin positive: arrows) are emerging to replace the oil droplets. This remodeling was progressing very slowly; the oil droplets stayed for a long time without scavenging or absorption and thus the adipose graft tissue can keep its volume even during the remodeling phase. This phenomenon cannot be detected by H&E staining, because it cannot discriminate alive adipocytes from dead ones.

FIG. 48.17 Summarized schema of adipocyte fate after nonvascularized adipose grafting.

Animal experiments of nonvascularized adipose tissue grafting indicated that most of the adipocytes (except for those in the most superficial layers) die within 2–7 days after adipose tissue grafting. Some of the dead adipocytes are replaced with new adipocytes of the next generation, while others are not; in the central area of the graft, even adipose stem/progenitor cells die and the replacement of new adipocytes fails, resulting in scar tissue formation or cyst formation with frequent calcifications in the end. The final volume retention after adipose grafting is determined by the rate of successful replacement of adipocytes.

Optimization of Individual Procedural Technique

If one of the four key procedures (liposuction, storage, preparation, and injection) as well as preoperative care was not properly performed, good clinical outcomes cannot be expected. Every step has been standardized, though the standardization is not yet completed and new techniques or devices continue to be developed. Aspirated adipose tissue is not vascularized and many adipocytes die every second, so that harvested adipose tissue should be kept cool and reinjected as quickly as possible after appropriate preparation. Fat grafts should be distributed into the proper tissue layers as diffusely as possible (as small aliquots) to maximize the contact surface area of grafts.

Key Factors of Graft Materials

For preparation of adipose grafts, we believe that there are two key factors that we should consider for a better retention rate.

• One is to improve the viable cell number per volume. This means concentrating needed components (viable adipocytes, ASCs, and other cellular components) and discarding unneeded components such as water, oil (ruptured adipocytes) and erythrocytes. This is usually performed through centrifugation or filtration with negative pressure (Fig. 48.7). Centrifugation concentrates cellular components and separates water and oil with variable degrees, depending on the centrifugation force and time.¹²

• The other key point is to improve the cellular quality of the graft. This means improving (normalizing) the progenitor (ASC)/adipocyte ratio; aspirated adipose tissue has been suggested to have a lower ratio than intact adipose tissue.⁵ In general, stem/progenitor cells and activating signals are crucial factors for tissue regeneration. In fat grafting, there are many signals to activate stem/progenitor cells such as injury or bleeding, but dying adipocytes seem to be crucial signals to induce ASC differentiation into adipogenic lineage. If the adipose graft did not contain enough stem/progenitor cells, the tissue regeneration would not be efficient. On the other hand, if only stem/progenitor cells are transplanted, stem cells would not be either activated or differentiated into the expected direction. So, if the tissue is stem-cell-deficient in number, it would be reasonable to normalize stem cell density in the tissue by supplementing the stem/progenitor cells as performed in CAL strategy (Fig. 48.2) or by reducing the adipocyte number. Centrifugation can relatively reduce the viable adipocyte number without losing ASC¹² (Fig. 48.7).