Autologous Fat Transplantation for Aesthetic Breast Augmentation: A Systematic Review and Meta-Analysis Free

Abstract

Background

Autologous fat transplantation has already become a part of clinical practice for aesthetic breast augmentation even though evidence regarding its efficacy is still lacking.

Objectives

The authors sought to determine the current worldwide status and efficacy, techniques, and oncologic safety on this subject.

Methods

PubMed, EMBASE, and Cochrane Library databases were searched to identify all relevant studies.

Results

Eighty-four articles published between 1987 and April 2020, consisting of 6468 patients, were included, and 64 studies consisting of 5162 unique patients were included in the meta-analysis. Most studies had a low level of evidence (levels 2b-5); In this meta-analysis, there were 17 prospective cohort studies, 4 retrospective cohort studies, 6 case-control studies, and 38 case series. The publications were from 21 countries. Indications for autologous fat transplantation were aesthetic augmentation (93.2%) and congenital malformation (6.8%). Among the 5162 patients, 2 cases (0.04%) of cancer were reported. The meta-analysis revealed very high overall patient and surgeon satisfaction rates of 93% and 87%, respectively. Overall, only 1.56 sessions were needed to achieve the desired result. Long-term survival was calculated to be approximately 60% to 70% at 1-year follow-up. Only 8% of procedures resulted in clinical complications, and 5% of patients required biopsy because of abnormal clinical or radiological findings.

Conclusions

Autologous fat transplantation seems to be a major tool in aesthetic breast augmentation. Preoperative patient selection is essential but under-reported. Future research should focus on evaluating the technical and patient factors influencing the rate of fat survival and its oncological safety.

Level of Evidence: 4

More than 100 years have passed since autologous fat transplantation (AFT) was utilized as a relatively simple and effective procedure for treating soft-tissue deficiencies. Neuber¹ first reported the operation of filling soft tissue defects with several small, free fat blocks in 1893. In 1895, Czerny² reconstructed the damaged breast with lipoma from the waist, which was the earliest record of breast augmentation with autogenous fat.

In 1987, Bircoll³ injected the breast with autologous fat grain for the first time. Since then, the application of AFT in breasts has been developing. Although AFT in breasts was forbidden by the American Society of Plastic Surgeons because of issues related to efficacy and safety, there is increasing interest in employing AFT for other indications. Thus, in 2009, AFT was not contraindicated for natural breasts but was regulated in the United States following the American Society of Plastic Surgeons Fat Graft Task Force report.⁴

AFT is utilized in congenital hypoplasia of the breast (unilateral or bilateral), such as micromastia, or congenital breast malformation (eg Poland syndrome, tuberous breast). It is also utilized for atrophic breasts after breast-feeding and residual local volume deficiencies after breast reconstruction following lumpectomy. Fat transplantation is an autologous reconstruction by a minimally invasive procedure compared with autogenous tissue, such as the myocutaneous flap.

Numerous studies^5-25 about the widespread application of AFT in the breast have been published over the years. However, most of them were the individual findings of primarily case series and cohort studies that differed in indications, surgical techniques, and outcome measures that were highly fragmented and lacked reliable data from randomized controlled trials (RCTs). Thus, there is an urgent need for a thorough examination of the published literature on AFT in the form of a meta-analysis to facilitate a more straightforward interpretation by clinicians and help reach consensus regarding the efficacy of AFT in aesthetic augmentation of the breast.

New technologies and related research are updated continuously in the aspects of fat absorption, purification, and injection to avoid all kinds of complications and improve the survival rate of autogenous fat. For example, breast augmentation with the body-jet hydrodynamic liposuction system,⁵ Brava-assisted augmentation mammoplasty,^6,7 cell-assisted augmentation,^8,9 and augmentation mammoplasty with autologous fat combined with a prosthesis^10,11 are recognized by an increasing number of patients and plastic surgeons.

This is the first review, to our knowledge, to determine the current worldwide status and efficacy, techniques, and oncologic safety of autologous fat transplantation for aesthetic breast augmentation.

METHODS

Search Strategy

This systematic review adhered to the standards of the Cochrane Handbook for Systematic Reviews of Interventions² and was written in the format provided by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement.³ A comprehensive, reproducible electronic search was conducted in the PubMed, EMBASE, and Cochrane Library databases to identify all published studies with human patients who underwent AFT for aesthetic breast augmentation. The search was performed on April 1, 2020, employing the following terms: “fat grafting,” “fat graft,” “fat transplantation,” “fat transfer,” “fat injection,” “fat implantation,” “fat filler,” “lipofilling,” “lipotransfer,” “lipomodeling,” “lipostructuring,” “lipofiller,” “lipoinjection,” “lipograft,” “adipocyte graft,” and “breast” (Details in Appendix).

Selection Criteria

We included all original articles that were published before April 1, 2020, concerning patients who underwent AFT for healthy native breasts, such as those with micromastia, Poland syndrome, tuberous breast deformity, and atrophic breast. The patients were required to have no personal or family history of breast cancer. Family was defined as first-degree relatives. The patients underwent only AFT or AFT combined with 1 of 3 auxiliary measures, including supplements (eg, cell-assisted lipotransfer, adipose-derived stem cells, platelet-rich plasma, and stromal vascular fraction [SVF]), preexpansion devices (eg, the Brava device), and implants. Exclusion criteria were unavailability of the full-text article, secondary sources (reviews and commentaries), and the utilization of AFT for breast reconstruction after cancer.

Data Collection

The full text of each article was read and reviewed by 2 independent reviewers (Y.J.W. and F.H.) based on predefined inclusion and exclusion criteria. Disagreements were resolved through discussion until a consensus was reached. First, we extracted general data from the literature with qualitative synthesis according to the predetermined table; the data information table included authors, publication date, study location, type of study, total number of patients, level of evidence, and average patient age. Second, basic data and quantitative data included treatment measures, body mass index (BMI), sessions of operations, follow-up duration, patient and surgeon satisfaction, complications, biopsies, mean fat volume injection, and volume retention. Third, perioperative information included the type of anesthesia, donor site, liposuction, fat treatment, fat injection, injection level, average operative time, and postoperative nursing. Fourth, postoperative follow-up included the type of imaging examination, radiological and oncological results, volume measurement method, time of breast volume change, and types and number of complications.

In addition, if studies from the same author or institution were conducted at the same time and reported the same outcomes, an overlap of more than 25% of the sample size was suspected. In case of the above situation, only the largest or most relevant study was included in our meta-analysis.

In some instances, authors were contacted for additional data. When necessary, units were converted to a standard format to ensure comparability and allow the pooling of data. Continuous variables that were reported as medians ± ranges were converted to means ± standard deviations employing the standard estimating equations utilized for meta-analyses. Similarly, categorical outcomes that were reported as Likert scale scores (eg, the degree of satisfaction) were dichotomized to allow analysis as proportions.

All references were stored in the Endnote Reference Management Tool (version X7.8, Thomson Reuters, Philadelphia, PA).

Outcome Effect Indexes

The combined effect indexes of a single-arm study included patient satisfaction, surgeon satisfaction, average sessions of operations, complications, and biopsy rate.

Statistical Analysis

Meta-analyses were performed with STATA/SE15.1 (TX77845, package meta; StataCorp, College Station, TX). To conduct the meta-analysis, the sample size of 1 arm was more than 4, and the performance of AFT had relevant outcome measures, including the rates of patient and surgeon satisfaction, volume retention, the procedure sessions, and associated clinical and radiological complications.

Heterogeneity tests were performed employing the I² statistic, and the method of combining the effect index was inverse variance. All of the effect indexes were pooled in a standard random-effects model and presented as forest plots. All of the combined effects are expressed as rates and 95% confidence intervals.

The bias within a single study may be an indication bias, so 3 selection criteria were utilized to reduce the risk of indication bias: (1) exclude individuals with a personal history or first-degree relative family history of breast cancer and individuals who may already be at risk for cancer; (2) select individuals with only 1 combination of interventions; and (3) select individuals who underwent AFT of the breast for the first time.

At the same time, to reduce the heterogeneity among the studies, subgroup meta-analyses were performed for most outcomes by segregating studies based on known or presumed confounders (eg, interference measures). Depending on the type of interference and according to the technique and method of AFT, the studies were categorized into 4 subgroups: pure AFT (AFT group), combined supplement (AFT + SUPP group), combined Brava expansion technique (AFT + BRAVA group), and combined implants (AFT + IMP group).

To investigate the source of heterogeneity, a meta-regression model was established. There were multiple covariates: postoperative follow-up duration, manual or mechanical liposuction, fat transplantation subgroups, and fat survival rate. The meta-regression model trendline and respective 95% confidence intervals were utilized to estimate fat survival rate and volume retention over time.

RESULTS

According to the systematic review selection criteria, the electronic search yielded 1749 articles, screening of related articles yielded 1 additional article, and 1087 articles were obtained after eliminating the duplicates. Screening of the title and abstract led to the inclusion of 131 records for further evaluation. Eighty-four studies were selected through further screening of the full-text articles. According to the meta-analysis selection criteria, 64 articles were finally included in the meta-analysis. The retrieval strategy flowchart is shown in Figure 1.

Study Characteristics

Eighty-four clinical studies, consisting of 6468 patients, were identified for inclusion (Table 1). Patient age ranged from 11 to 63 years, and the average age was 35.73 years. These studies were conducted between February 1987 and April 2020, mostly after 2008 (Figure 2). The studies were from 21 countries, with most being from North America, Europe, and Asia (Table 2). Most of these studies had a low level of evidence (levels 2b-5). Only 1 study was a randomized controlled cohort study, but the control groups were low-speed centrifugation group and sedimentation group according to different fat processing. Our control standard was according to the technique and method of AFT reinjection, and the remaining AFT arms were treated as case series. Sixty-four studies comprised 5162 unique patients who were included in the meta-analysis (Table 3). These studies consisted of 38 case series, 6 case-control studies, 17 prospective cohort studies, and 4 retrospective cohort studies. The average follow-up duration was 22 months (range, 3 months to 25 years). Patients’ average BMI was 17 to 30 kg/m². The indications for AFT were aesthetic augmentation in 93.2% of patients (n = 4814) and congenital malformations (Poland syndrome and tuberous breasts) in 6.8% (n = 348).

Perioperative Management

Details of the studies are listed in Table 4.

Anesthesia

In most studies, surgery was performed with the patients under general anesthesia combined with local infiltration anesthesia. In some cases, only local infiltration anesthesia, namely tumescent technology, was employed. Ingredients were 30 to 50 mL of 1% lidocaine and 1 mL of 1:100,000 adrenaline in each liter of normal saline or sodium lactate format solution. The infiltration lasted for 10 to 15 minutes. However, lidocaine was not present in the tumescent fluid in 9 studies (833 patients). Five studies (373 patients) utilized antibiotics after the operation, and only 2 studies detailed the method of antibiotics.

Brava Device

Pre-expansion with the Brava device was employed in 7 studies (779 patients). The main objective was to improve the surfaces that interfere with AFT through neovascularization by pre-expanding the receiving tissues. The method involved wearing the Brava device for 3 to 4 weeks before the operation for 10 to 12 hours per day and 2 to 4 weeks after the operation.

Preparation of the Breast

Only 2 studies mentioned the preoperative preparation of the breast area. Ho Quoc et al¹² utilized a 14-gauge trocar to perform fasciotomies of the breast area so the fat could survive more easily under less pressure. Zocchi and colleagues⁷ pressurized inflation with carbon dioxide in the breast area.

Donor Areas

The typical fat donor areas employed were the abdomen and thighs, but this varied depending on both the individual patient fat-tissue distribution and where the surgeon could most easily access the donor sites.

Fat Harvesting

The principle of liposuction is based on the Coleman technology,¹³ 2- to 4-mm-diameter casing, and negative 0.4 to 0.6 standard atmospheric pressure. However, there were some small differences among the studies, such as the negative pressure source, including a manual syringe and machine. Twenty-two studies performed liposuction with the manual syringe, 37 studies with a machine, and 7 studies (in 37 studies) utilized a hydrodynamic liposuction machine.

Fat Processing

Fat tissues were usually processed by centrifugation in 37 studies. The optimal rotation speed and time were not provided, but most of them reported values of 3000 rpm and about 2 to 4 minutes, respectively. Sedimentation was employed in 12 studies, filtration in 3, and the cleaning method in 4. Kang and Luan¹⁴ reported the effect of low-speed centrifugation and sedimentation on postoperative fat necrosis. This study suggested that low-speed centrifugation was more likely to cause refractory nodules, especially for patients who have had breast surgery or need multiple fat transplantations. Twelve studies utilized SVF or adipose stem cells derived from autogenous fat, one of which reported a case of auto-concentration of monocytes derived from bone marrow obtained from the ilium. Peltoniemi et al¹⁵ and Chiu¹⁶ reported no significant difference in the fat survival rate between the AFT group and AFT combined with SVF at 6 and 12 months postoperatively. Dos Anjos et al¹⁷ categorized SVF into a low-concentration group and a high-concentration group for comparison, and the fat survival rate at 18 months was significantly different between the 2 groups (50% vs 75%).

Fat Reinjection

Multi-tunnel, multi-layer, multi-point, fan-shaped, and small have been commonly employed to describe the fat injection. Syringes range from 1 mL to 60 mL in capacity. A spiral-propelling syringe was also utilized, and each spiral rotation injected 0.28 mL of fat. Most studies injected fat in the hypodermic, hypoglandular, upper, and lower spaces of the pectoralis major muscle. However, it is controversial if the injection level in the pectoralis major muscle causes intramuscular nodules and pain during activity. Different researchers have different opinions on the amount of injection to employ. The amount of a single injection in the AFT group was 50 to 420 mL, that in the AFT + SUPP group was 195 to 380 mL, that in the AFT + BRAVA group was 245 to 430 mL, and that in the AFT + IMP group was 27 to 134 mL. Five studies^18-22 proposed that because fat tissue must be absorbed after the operation, the amount of transplanted fat should be 30% to 50% more than the expected amount. Thus, the amount of relative stable fat survival at the final 12 months minimum was appropriate and satisfactory. Longer term satisfaction still needs to be observed. In addition, 2 studies proposed that the interval between 2 operations was 3 months or 6 months, which was consistent with our results in the meta-regression model established with postoperative follow-up duration as the covariate and fat survival rate as the dependent variable. The sessions of operations were usually approximately 1 to 5 times for achieving better results.

Postoperative Follow-up

Details of the postoperative follow-up for all studies are listed in Table 5.

Imaging Examination

Almost all of the studies reported 1 or more imaging examinations utilizing ultrasonography, mammography, and magnetic resonance imaging (MRI). Patients underwent preoperative and/or postoperative radiographic imaging. Mammography was the most common, with 24 studies reporting abnormalities, most of which were cysts and calcifications. Eight studies were conducted in accordance with the American College of Radiology Breast Imaging Reporting and Data System (BI-RADS) standard mammography evaluation, most of which were BI-RADS 1 or 2 before the operation or BI-RADS 2 or 3 after the operation. In the control examination after 6 months, most of the mammograms classified as BI-RADS 3 were reclassified as BI-RADS 1 or 2. Spear and Pittman²³ reported a case of BI-RADS 4 that showed a negative pathological finding after biopsy.

Regarding the assessment of volume gain, most of the studies mainly reported MRI (24 studies) and 3-dimensional (3D) laser scanning (7 studies), which were performed at 3, 6, and 12 months postoperatively. In addition, some other methods were utilized: 5 studies measured the chest circumference, 2 studies measured the distance from the sternum to the inner edge of the breast, 1 study measured the thickness of the breast, and 2 studies employed digital software imaging.

Complications

In the 53 studies retrieved, complications were identified in approximately 8% of patients (388/5162). The main complication was oil cysts. Other common complications were calcification, fat necrosis, small nodule, infection in the receiving area, asymmetry, insufficient transplantation, depression in the donor area, and hematoma. Pneumothorax was rare and occurred in only 1 case. Thirty patients required further surgical treatment.

Oncological Risk

Of 5162 patients followed-up for an average of 22 months (range, 5-72 months), 2 were diagnosed with breast cancer (0.04%). The 2 cases were reported by Coleman and Saboeiro¹³ in a case series of 15 patients published in 2007, with cancer in the non-transplant area in 1 case and cancer possibly in the transplant area in another case.

Meta-analysis

Patient and Surgeon Satisfaction

The data of 4413 patients in 29 studies were converted by double anti-sine and combined with the random effect model employing the D-L method. After a mean follow-up of 22 months (range, 6 months to 10 years), the meta-analysis revealed an overall patient satisfaction of 93% (95% CI = 89%-97%) (Figure 3). According to the heterogeneity test, there was no significant difference between the subgroups (P > 0.01). Similarly, the satisfaction of surgeons was 87% (95% CI = 80%-94%) in 742 patients from 11 studies (Figure 4).

Complications

The incidence of AFT-related clinical complications based on our meta-analysis was 8% (95% CI = 5%-11%) in 53 studies (Figure 5). There was no significant heterogeneity between the subgroups (P > 0.01).

Number of AFT Sessions

The meta-analysis of data from 2180 patients from 36 studies provided an overall mean number of AFT sessions of 1.56 (95% CI = 1.26-1.85) (Figure 6). The test result of heterogeneity between the subgroups was I² = 98.8% (P < 0.01). There was significant heterogeneity between the groups, which may be related to the implementation of surgery by different surgeons, patients’ conditions, and perioperative conditions.

Autologous Fat Transplantation-Related Biopsy Rate

The final meta-analysis showed that 5% (95% CI = 2%-12%) of breasts treated with AFT in 12 studies (670 patients) had to eventually undergo additional histopathological examination because of the presence of abnormal radiological images or a clinically detected palpable mass (Figure 7). Among the 30 cases of biopsy, 28 were found to be poor on pathological examination, and 2 were considered breast cancer cases.

Meta-regression Model

Among the results of 21 studies (1483 patients), 3 factors, including the postoperative follow-up time, manual or mechanical liposuction method, and 4 subgroups of fat transplantation methods, were employed as covariates, and the fat survival rate was utilized as the dependent variable to establish the regression model (P < 0.01). The model suggested that there was heterogeneity among the studies in postoperative follow-up duration, especially at 12 months postoperatively (Table 6). There was no heterogeneity (P > 0.01) between the studies that employed the liposuction and fat transplantation methods (Table 7).

Twenty-one studies provided data on volume retention, which was plotted against time in a meta-regression model (Figure 8). The meta-regression model trend line was 60% to 70% at 1-year follow-up and reached a plateau at approximately 65% when extrapolated on the long term. Only 1 study (57 patients) provided follow-up data at 18 months, and this had a great influence on the trend-line.

DISCUSSION

Synthesis of Results

Over the past decade, there has been renewed interest in AFT for breast augmentation. Instead of focusing on the application of AFT for breast reconstruction, research has gradually shifted to aesthetic breast augmentation, and many studies have been performed. This systematic review and meta-analysis of 6468 cases from 84 studies of aesthetic breast augmentation by AFT in the past 33 years provides a more sophisticated and balanced interpretation of study evidence on AFT to allow healthcare providers to make evidence-based recommendations on the efficacy of the technique.

Sixty-four studies reporting 1 or more relevant outcomes reflecting the efficacy of AFT were selected for this meta-analysis. Patient satisfaction was the most commonly reported outcome, and it was 93% in 4413 unique patients from 29 studies. Similarly, 87% of plastic surgeons were pleased with the result based on preoperative and postoperative photographs. Other evaluation methods of breast augmentation were the satisfaction rating, Likert scale, visual analog scale, and Breast-Q. Only 5 studies^11,21,23-25 employed the Breast-Q to investigate patient satisfaction (158 patients), which involves the evaluation of breast shape, size, and texture as well as the quality of life. Volume retention was evaluated in 1483 patients from 21 studies by performing serial measurements utilizing 3D imaging, MRI, or computed tomography. In the past, only MRI and computed tomography could achieve such measurements, but their utilization was limited because of their high cost and high radiation exposure. The recent development of 3D imaging equipment made these measurements more practical, and it is the most suitable and reliable method for correcting insufficient volume in AFT. Volume retention results varied greatly between studies, but the meta-regression model against time showed 50% volume retention at 1 year. Unfortunately, the data were severely limited, because measurements beyond 6 and 12 months were lacking. Moreover, the possible effects of other important confounders such as fat harvesting, fat processing, and fat reinjection could not be demonstrated until now. Interestingly, the overall number of sessions required was relatively low (1.56). Although it can be argued that this number would become higher if excessive fat resorption would result in patients demanding extra AFT sessions, the mean follow-up duration of 22 months would be sufficient to accommodate for this. Unfortunately, the rate of clinical complications per procedure was 8%, and these complications were not minor. Finally, the risk that a breast treated with AFT would result in clinical or radiological abnormalities and require histopathological examination (biopsy) was 5%. Two cases of cancer diagnosis were reported among the 5162 patients (0.04%). At present, there is no special study on tumor oncological safety of AFT to our knowledge, but these cases were not representative of the whole population. Further study with long-term follow-up is needed.

Limitations

This meta-analysis has a couple of limitations. First, there was a low level of evidence due to the lack of RCTs. AFT has become a part of clinical practice, and it may be considered unethical to establish an RCT with a standard blind method and control group. In addition, because of the lack of appropriate and safe alternative methods, few studies have successfully included a control group. Therefore, data collected over the past 30 years were sufficient for a single-arm meta-analysis. By determining the size, direction, and CI of each result, the comprehensive effect evaluation of a large number of observational studies is mainly provided. Second, standard and validated measurement results tools (eg, Breast-Q) were rarely employed to improve the effectiveness and comparability of different studies. Over the years, researchers have relied on various unverified, customized scales, and questionnaires or utilized preoperative and postoperative photographs. Third, the follow-up duration was too short. If examining cancer and outcomes of AFT, 2 years would be the minimum follow-up. In addition, the results of effect indicators are often regarded as continuous data (mean ± standard deviations) or classified variables (frequency). The main problem with continuous data is that there is no direct reference value or comparison between the control group and the combined effect results. Even if these data are available, it is difficult to assess whether the overall data have the same distribution. However, classification methods (eg, dichotomies) tend to oversimplify results by assigning patients to specific categories (dissatisfaction and satisfaction). Because these data do not depend on the reference value, they can more accurately reflect the absolute effect of autogenous fat transplantation, but they lack the accuracy of continuous data. As with all meta-analyses, heterogeneity between studies can have an important effect on the results. Therefore, all of the analyses in this study adopted the random effect model and analyzed different interference measures in the subgroups to consider the heterogeneity between the studies.

Future Recommendations

This meta-analysis provides strong evidence for the overall efficacy of AFT in breast augmentation and establishes a foundation for further research in this field. A specific subgroup analysis will improve the quality of future research methods. Such future studies should be conducted employing prospective study designs, including appropriate control groups and the utilization of validated outcome measurement instruments or standards. According to the current data, in the next 5 years, the global plastic surgery community should solve the following research problems. First, clarify the long-term oncological safety of AFT. Second, determine the best AFT technology that can predict the long-term survival rate. Third, determine the factors that affect the survival rate, such as height, chest circumference, and BMI, and the relationship between them as well as the maximum amount of AFT.

CONCLUSIONS

After an average of 1.56 operations (range, 1-5), patient satisfaction (93%) and surgeon satisfaction (87%) reached high levels, indicating that AFT has a significant effect on breast volume and contour improvement in breast augmentation. The volume retention after 1 year was 60% to 70%; however, further research is needed to confirm this finding. Mild complications occurred in 8% of patients, and biopsies were performed in 5% of patients. AFT can be employed as one of the methods of breast augmentation. However, preoperative imaging examinations, such as breast ultrasonography, mammography, or MRI, should be performed. The indications of AFT should be clear, the operation should be performed carefully and in a standardized manner, complications should be avoided as much as possible, and long-term follow-up should be conducted after the operation to ensure the oncological safety of AFT.

Acknowledgments

We thank Editage (www.editage.cn) for English language editing.

Disclosures

The authors declared no potential conflicts of interest with respect to the research, authorship, and publication of this article.

Funding

The authors received no financial support for the research, authorship, and publication of this article.

Availability of Data and Materials

The datasets generated in this study are available from the corresponding author on reasonable request.

REFERENCES

Author notes