According to statistical data from The International Society of Aesthetic Plastic Surgery in 2021, breast augmentation will continue to be the most commonly performed surgical intervention among women [1]. In recent years, autologous fat grafting has gained significant attention as a substitute for conventional breast implantation. Nevertheless, concerns persist regarding the inherent unpredictability and suboptimal graft survival rate of this procedure, warranting further investigation and evaluation [2-4]. Women with a body mass index (BMI) <18.5 kg/m² are classified as underweight [5], which often raises concerns about their limited fat reserves, making them potential contraindications for autologous fat grafting in breast augmentation. Many women seeking breast augmentation have underdeveloped breasts and are categorized as underweight. In this retrospective study, we examined patients with a BMI <18.5 kg/m² to assess the efficacy and safety of autologous fat grafting for breast augmentation.

Between May 2019 and May 2021, 390 patients underwent autologous fat grafting for breast augmentation at our clinic. Among these patients, 63 were underweight with a BMI of <18.5 kg/m². After excluding individuals who underwent multiple treatment sessions and those lost to follow-up, a final cohort of 53 patients was included for analysis in this study.

Indications for this procedure include cosmetic augmentation, correction of contour deformities following implant removal, and correction of congenital breast asymmetry. Prior to the procedure, all patients received comprehensive information regarding potential complications and provided written informed consent as a testament to their understanding and agreement.

Standardized protocols were employed for physical examination and breast ultrasonography at 3 and 12 months post-treatment to assess and identify potential complications. Comprehensive clinical data regarding all observed post-treatment complications were meticulously recorded during the follow-up period for all patients. In cases where a palpable mass or abnormal findings were detected during physical examination or ultrasonography, patients were promptly referred to the hospital for further evaluation using magnetic resonance imaging (MRI).

This study was approved by the Institutional Review Board (IRB 22-001G-1) at the Genesis Clinic and was conducted in compliance with the 1975 Declaration of Helsinki. All patients provided written informed consent for the procedure and had been advised of the potential complications of fat grafting to the breast.

Harvesting of adipose tissue

Potential donor sites for autologous fat grafting in underweight women included the abdomen, flanks, hips, thighs, upper arms, and calves. It is noteworthy that underweight women typically exhibit a relatively limited subcutaneous fat layer compared to those with normal weight. Therefore, the suitability of each potential donor site for liposuction should be assessed thoroughly. Before the fat harvesting procedure, all patients underwent intravenous sedation and local tumescent anesthesia. To ensure optimal conditions, each harvest site was infiltrated with 150–300 ml of tumescent anesthesia consisting of 1,000 ml of lactated Ringer’s solution, 40 ml of 2% lidocaine, and 1 ml of 1:1,000 epinephrine, administered 10 minutes before initiating liposuction.

Third-generation ultrasound-assisted liposuction (UAL) was performed using an Ultra-Z system (Zerone Co., Ltd., Seoul, Korea). A 3.7-mm, 3-ring probe was employed at an amplitude of 100% in normal mode (10 Hz) for the donor sites. Subsequently, the subcutaneous fat was emulsified, and adipose tissue was harvested using a -3-mm or 4-mm aspiration cannula connected to a low-pressure suction machine set to approximately -500 mmHg. After fat extraction, we followed Coleman’s recommended purification method, which involves centrifugation at 1,200G for 3 minutes to minimize damage to the fat tissue.

Injection technique

Injections were administered to patients in the supine position under intravenous sedation. Following the principles of the Coleman method, injections were administered using a 14G cannula measuring 15 cm in length. Injections were administered in a fanning pattern and in small aliquots during withdrawal. Moreover, a specific strategy known as the “solid injection technique,” previously described in the author’s published article, was closely adhered to [6-8]. Two entry points were used for the injections: one located at the inframammary fold and the other at the medial periareolar area.

Volumetric analysis

Objective volume calculations of the breasts were performed using a noncontact three-dimensional (3D) laser surface scanning technique with a portable device; the Konica Minolta Vivid 910 3D Digitizer (Konica Minolta Inc., Tokyo, Japan). The scanning procedure, which involves capturing multiple views for merging, was completed in less than 60 seconds [9]. The data acquired from these scans were then merged for volumetric analysis using “Rapidform XOV2” software (INUS Technology, Inc., Seongnam, Korea) on a computer. Volumetric analysis was conducted for each breast in all patients by an expert blinded to the treatment both before treatment and at the 3- and 12-month follow-up assessments.

Subjective evaluation

One year postoperatively, patient satisfaction was evaluated using a 5-point Likert scale. The scale ranges from 1 (very dissatisfied) to 5 (very satisfied). The satisfaction level was determined by calculating the proportion of patients who rated their satisfaction as 4 or 5, representing those who were satisfied or very satisfied, respectively.

The average age of the participants enrolled in the study was 31.6 years (standard deviation [SD]=7.6). Their average BMI was 17.7 kg/m² (SD=0.7), and the mean percentage of body fat was 21.4% (SD=3.1). The average volume of aspirates obtained during the procedure was 1,410 ml (SD=90), and each breast received an average injection volume of 295 ml fat (SD=47). The mean follow-up period was 37.9 months (SD=7.4). During follow-up assessment, routine postoperative breast ultrasonography revealed 2 cases (3.7%) of fat necrosis. The mean volume retention of the grafted fat assessed by noncontact 3D laser surface scanning was 69.9% (SD=4.2) at the one-year postoperative mark (Fig. 1). Additionally, the overall patient satisfaction rate was remarkably high (94.3%). Table 1 displays the distribution of patient data and provides an overview of the study population. The fat extracted from the UAL and comparison images of the patient’s donor site before and after liposuction are shown in Fig. 2; Fig. 3–5 depict the surgical outcomes of the patients and provide visual representations of the results.

Table 1 . Patient demographics.

Characteristics	Total number (n=53)
Age (yr)	31.6±7.6
BMI (kg/m2)	17.7±0.7
Fat ratio (%)	21.4±3.1
Aspirate (ml)	1,410±90
Graft injected per breast (ml)	295±47
Follow-up time (mo)	37.9±7.4
Graft retention (%)	69.9 (4.2)
Complications (%)	2 (3.8)
Patient satisfaction rate (%)	50 (94.3)

Values are presented as mean±standard deviation or number (%)..

BMI, body mass index..

Figure 1. Preoperative and postoperative breast volumes were assessed utilizing a noncontact three-dimensional (3D) laser scanner (Minolta Vivid 910 3D Digitizer; Konica Minolta Inc., Tokyo, Japan) equipped with a customized portable device (A). The 3D laser scanner captured the breast surface, generating a merged 3D image for each patient (B).

Figure 2. Fat extracted after ultrasound-assisted liposuction, showing emulsified adipose tissue and less blood-tinged fluid compared to traditional liposuction (A). Purified fat awaiting injection (B). Patient 3 underwent ultrasound-assisted liposuction with the extraction of 1,400 ml of fat from the circumthighs and knees. Preoperative anterior and posterior views of the thighs (C, E) and postoperative 3-month anterior and posterior views of the thighs demonstrate a smooth surface of the liposuction area without any irregularities or depressions (D, F).

Figure 3. Patient 1 is a 35-year-old female with a body mass index of only 18 kg/m² (height: 158 cm, weight: 44.9 kg) and a body fat percentage of 20.3. She has never had children and her breasts are very flat with no distinct inframammary fold (A, C, E). We extracted a total of 1,520 ml of fat from her inner thighs, anterior thighs, outer thighs, and waist. After centrifugation and purification, we injected 330 ml of autologous fat into her right breast and 320 ml into her left breast. One year later, upon follow-up examination, we observed a deepening of her cleavage, a significant increase in breast projection, an added thickness to the north pole, and a remarkably noticeable change in appearance (B, D, F). The patient expressed great satisfaction with the outcome.

Figure 4. Patient 2 is a 37-year-old female who has given birth to two children, with a body mass index of 18.4 kg/m² (height: 160 cm, weight: 47.1 kg) and a body fat percentage of 24.8. Her breasts have experienced significant postpartum involution, and the breast skin appears loose and sagging like a bag. The nipple areolar complex is distorted and drooping (A, C, E). We extracted 1,660 ml of fat from her inner thighs, outer thighs, anterior thighs, and upper outer buttocks. After centrifugation and purification, we injected 400 ml of autologous fat into the right breast and 340 ml into the left breast. One year later, during follow-up, we observed a significant increase in breast volume, deepening of the cleavage, enhanced projection, and aesthetically pleasing appearance of the breasts. The nipple areolar complex also improved in appearance (B, D, F).

Figure 5. Patient 3 is a 26-year-old female with a body mass index of 17.3 kg/m² (height: 160 cm, weight: 44.4 kg) and a body fat percentage of only 18.9. She has severely underdeveloped breasts, with a flat sternum area and no inframammary fold (A, C, E). We extracted a total of 1,400 ml of fat from her anterior thighs, posterior thighs, inner thighs, outer thighs, and knees. After centrifugation and purification, we injected 280 ml of fat into the right breast and 280 ml into the left breast. One year later, during follow-up, we observed the formation of a cleavage, increased breast volume, and the presence of a new inframammary fold in the sternum area (B, D, F). Although the patient initially desired larger breasts, she expressed great satisfaction with the outcome.

Underweight women typically have a lower body fat percentage, which presents challenges in obtaining an adequate amount of fat for grafting [10]. Additionally, in extremely thin women who have never been pregnant, the tightness of the breast skin and narrowness of the breast space greatly limit the volume of autologous fat that can be injected. However, our study primarily focused on underweight women; therefore, we did not specifically analyze differences before and after childbirth. Regarding women who have undergone childbirth, as in patient 2, it is acknowledged that the breast tissue tended to be more lax, which may result in decreased surgical complexity.

Some physicians have suggested that women with insufficient fat should gain weight before undergoing autologous fat grafting [11]. If patients increase their body fat percentage prior to surgery, it would be easier to extract an adequate amount of fat. However, this is not advisable. If the patient experiences weight loss after surgery, the breasts may significantly decrease in size, leading to disappointment for the patient.

Therefore, patients were not encouraged to gain weight prior to surgery. In contrast, we encouraged the patients to undergo the procedure at their original lower body weight. Normally, body weight tends to increase with age; as a result, the volume of autologous fat transplanted into the breasts also increases [12], leading to breast enlargement. If satisfactory autologous fat grafting for breast augmentation can be achieved in extremely thin individuals, their breasts will appear even larger as they gain weight owing to natural age-related changes. This is likely to result in a higher level of patient satisfaction.

In individuals with normal body weight, approximately 1,200 ml of aspirate is typically obtained from the inner, posterior, and lateral aspects of the thighs. However, in underweight individuals, acquiring this volume of adipose tissue may require harvesting from supplementary regions, including the abdomen, calves, and/or upper arms. Skillful surgical techniques are crucial to minimize deformities at all donor sites. In our experience, the 3 most common areas for liposuction are the inner, anterior, and lateral aspects of the thigh.

We overcome the challenge of low body fat percentage by utilizing UAL. Third-generation UAL for superficial fat extraction minimizes the risk of surface irregularities [13]. Even in areas such as the arms and calves, where deep fat is limited, this technique can be employed to emulsify fat and evenly extract it using negative pressure, thus avoiding the common risk of uneven contouring associated with traditional liposuction. Ultrasound energy can stimulate collagen production in the skin, leading to tighter skin 3 months after the procedure. Studies conducted on both humans and xenograft subjects have provided evidence that fat harvested through UAL exhibits stromal vascular fraction counts and graft retention comparable to those of traditional liposuction methods [14]. It has been established that adipose tissue obtained using third-generation UAL is viable at the time of harvest and has the potential to serve as a suitable source for autologous fat grafts. Schafer et al. [15] suggested that UAL is an efficient technique for harvesting adipose tissue without compromising the viability. The authors concluded that their findings aligned with those of previous reports demonstrating the clinical success of third-generation UAL.

The condition of the symmetrical areas was evaluated before fat extraction. If there was no asymmetry in fat thickness between the left and right sides, an equal volume of tumescent solution was injected into these areas. After emulsifying the fat using UAL, an equal volume of fat was extracted using negative pressure. After pre-emulsifying the fat with ultrasound, the ratio of water to fat in the extracted fat was almost identical, and the proportion of fat was very similar in the same volume of aspirate. As a result, there were minimal issues with the asymmetry between the left and right sides after surgery.

According to the literature, tight chest skin or insufficient breast space can be overcome by using recipient site pre-expansion with the Brava device (Brava LLC, Miami, FL, USA), followed by autologous fat grafting to the breast after a period of time [16-18].

However, this external expansion requires patients to carry a large hood on their breasts and a vacuum pump for weeks or months. Not only does it carry the risk of complications such as skin blisters and pigmentation, but it also imposes significant lifestyle inconveniences [19]. This is particularly unacceptable for Asian women, as the majority prefer not to publicly disclose their intentions to undergo autologous fat grafting for breast augmentation to friends or family members.

We employed the injection method known as the “solid injection technique,” as described by the authors in previously published literatures. This technique enables injection of the maximum amount of fat within minimal space, thereby enhancing fat viability and minimizing complications [6-8].

This technique involves 4 key principles. Principle I: Injection of fat into areas providing “solid” feedback during the injection process, with emphasis on using the non-dominant hand of the surgeon to enhance tactile feedback. This approach facilitates increased contact between host tissues and the newly grafted fat, leading to a reduction in the occurrence of liponecrotic cysts. Principle II: Avoiding injection of fat into “empty areas” characterized by a lack of feedback sensation, as this indicates placement of the cannula tip within a region abundant in the grafted fat. Further fat injection into these areas may lead to undesirable lumping. Principle III: Ensuring breast softness postinjection by preventing increased intramammary pressure resulting from improper fat placement. The “solid injection technique” focuses on delivering fat within the recipient tissue, minimizing pressure within the breasts. Principle IV: Prevention of pressure leakage from injection sites, which is considered unfavorable in autologous fat grafting for breast augmentation. Favorable outcomes and higher patient satisfaction were more commonly observed when the breasts remained soft and leakage-free.

In summary, the findings of this study indicate that a BMI <18.5 kg/m² is not a contraindication for autologous fat grafting in breast augmentation among women. With appropriate strategies, it is still possible to obtain sufficient fat in very lean individuals. Additionally, precise injection techniques allow adequate fat placement within the tighter breast skin and smaller breast spaces. Based on our results, the survival rate of transplanted fat can reach approximately 70%, with complications occurring within an acceptable range. Moreover, the patients expressed high levels of satisfaction with their outcomes.

Our study had several limitations. According to literature, postoperative satisfaction after breast surgery can be assessed using the BREAST-Q questionnaire [20]. However, owing to limited manpower, we used a simplified 5-point Likert scale to evaluate patient satisfaction. Additionally, considering economic and practical considerations, we did not routinely employ MRI to calculate the survival rate of the transplanted fat after grafting surgery.

The author has nothing to disclose.