Introduction

Among all of the innovative techniques available to provide breast restoration after mastectomy, most patients elect to pursue implant-based breast reconstruction (IBBR) [1,2,3]. This can be accomplished with two-stage reconstruction with tissue expander (TE) followed by exchange for definitive implant or with direct-to-implant reconstruction in a single stage, the former being the most popular in the U.S.A. [1,2,3]. By definition, a two-stage reconstruction requires an additional planned operation in which a permanent implant is placed after the expansion process has completed. During exchange, surgeons can not only modify the conus of the breast by inserting a larger or smaller sized implant but can also revise the breast footprint and envelope with a variety of procedures to optimize the esthetic outcomes of the reconstruction (e.g., capsulorrhaphy, dog ear excision, capsulectomy, advancement of capsular flaps, nipple reconstruction, etc.).

Fat grafting, also known as lipotransfer, is frequently performed as an adjunct to enhance the esthetic results and tissue quality of breast reconstruction [4]. In recent studies, fat grafting has shown the potential to reverse fibrosis secondary to postmastectomy radiation (PMRT) [5], and to improve sensation and pain when treating capsular contracture [6]. In the appropriate patients, simultaneous fat grafting during TE-to-implant exchange (SFG + TtIE) may be attempted in order to achieve optimal results more expeditiously, thereby avoiding revision procedures and their associated risks. SFG + TtIE incorporates a three-dimensional approach for the arrangement of the injected fat. In this setting, the mature capsule formed around the TE plays an important role as it defines the compliance and elasticity of the internal limit of the mastectomy flap [4].

Augmentation of the mastectomy flaps with autologous fat is usually achieved with various sessions of fat grafting after exchange with minimal complications [4]. However, when performed during TE-to-implant exchange, the amount of fat lipoinjected can increase the pressure within the mastectomy flap, compromising the perfusion to the periphery of the breast or incision site. Therefore, it is possible that SFG + TtIE may alter the ability of tissue to heal or increase the risk of infection after permanent implant placement. SFG + TtIE is not an uncommon procedure. Nonetheless, previous reports have not evaluated the impact and rate of complications associated with lipofilling of the mastectomy flap at the time of exchange. The aim of this study was to compare the rate of complications and the rate of additional fat grafting procedures between reconstructions that underwent SFG + TtIE and reconstructions that did not undergo fat grafting at the time of exchange (No-FGX).

Patients and Methods

Study Design

After approval by the institutional review board at a single large academic medical center, a retrospective review was performed of all consecutive patients undergoing two-stage IBBR with TE from January 2011 to December 2020. Patients who did not undergo TE-to-implant exchange or abandoned IBBR during the first stage were excluded. Patients who underwent hybrid immediate reconstruction with autologous tissue during TE insertion or during TE-to-implant exchange were excluded (e.g., implant-enhanced latissimus dorsi flap). Patients undergoing direct-to-implant reconstruction and patients who had incomplete data were also excluded.

Patients were subsequently divided into two groups by the surgical method implemented during TE-to-implant exchange (Fig. 1): The intervention group comprised reconstructions that underwent SFG + TtIE, while the control group comprised reconstructions that did not undergo fat grafting during exchange (No-FGX). The decision to perform either technique was based on the surgeon’s preference and after counseling with the patient. When an acellular dermal matrix (ADM) was used, the TE was either completely wrapped for a prepectoral device placement or partially wrapped along the inferior portion for submuscular reconstruction [7]. The surgical technique for a subpectoral or prepectoral approach has been previously reported [3, 7,8,9].

Fig. 1
figure 1

Diagram of the surgical management for SFG + TtIE in comparison with No-FGX. (IBBR, implant-based breast reconstruction; SFG + TtIE, simultaneous fat grafting during TE-to-implant exchange; TE, tissue expander; No-FGX, TE-to-implant exchange without simultaneous fat grafting)

Full size image

For SFG + TtE, we proceeded with fat harvest via suction lipectomy after tumescent infiltration. Fat was collected into injection syringes. Once the TE was explanted, capsulotomies, advancement of capsular flaps, and/or capsulorrhaphy sutures with PDS were performed as necessary. Sizers were opened and placed into the pocket. If the breast size was deemed appropriate with an optimal shape, contour, and good symmetry, fat was injected in the marked areas. Sizers were removed and the pockets were once again irrigated with antiseptic solution. The final implants were placed into the breast pockets afterward.

Study Variables

We recorded the following demographic and clinical variables of interest: the patient’s age at the time of reconstruction; body mass index (BMI); race/ethnicity; history of hypertension and diabetes; smoking status; breast cancer diagnosis and staging, including tumor and node status; estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) status; follow-up; history of breast irradiation and chemotherapy; indication of mastectomy (therapeutic/prophylactic); type of mastectomy (nipple-sparing mastectomy, NSM; skin-sparing mastectomy, SSM); laterality of breast reconstruction (right/left); type of procedure (bilateral/unilateral); timing of TE placement (immediate/delayed); weight of mastectomy specimen; plane of implant placement (prepectoral/submuscular); use of ADM; time to reach TE-to-implant exchange; final TE volume; and time from last outpatient device expansion to definitive implant placement.

Intraoperative and postoperative outcomes of interest were documented as follows: size of the definitive implant; implant–TE volume difference (Δ); surface of the definitive implant (smooth/textured); capsule work during TE-to-implant exchange (capsulotomy/partial capsulectomy/ total capsulectomy); and volume of fat lipoinjected during TE-to-implant exchange and after definitive implant placement. Complications recorded include hematoma, seroma, fat necrosis, leakage, infection, infection-associated implant removal, implant malposition, implant malposition requiring capsulorrhaphy, wound complications (mastectomy flap necrosis or dehiscence), wound-related prosthesis removal, capsular contracture, capsular contracture requiring capsulotomy or capsulectomy, requirement of autologous tissue for reconstruction salvage, explantation of prosthesis for all causes, and failure of IBBR. Failure of IBBR was defined as abandonment of any reconstructive efforts with prosthetic devices. Explanted implants secondary to a complication were regarded as implant removal, even if the implant was replaced during delayed procedure or if it was immediately exchanged. Infection was considered when (i) the reconstruction exhibited local inflammatory signs and per os or intravenous antibiotics were prescribed or (ii) evidence of a positive culture.

Propensity Score Matching

To optimize comparability between the intervention and control group, we performed one-to-one propensity score matching using multivariate logistic regression with a nearest-neighbor caliper width of 0.02. The propensity score was defined as the probability of undergoing SFG + TtIE versus No-FGX (dependent variable). We incorporated the age, BMI, smoking status, history of diabetes or hypertension, type of mastectomy (NSM/SSM), plane of implant placement (prepectoral/subpectoral), use of ADM (yes/no), capsule work during TE-to-implant exchange, implant surface (smooth/textured), adjuvant radiotherapy, and adjuvant chemotherapy in the propensity score model as independent variables.

Statistical Analysis

All breasts were treated as individual reconstructive cases; therefore, bilateral mastectomies in a single patient were regarded as two separate reconstructions. All statistical analyses were conducted with R statistical software, version 4.0.0 R Core Team [10]. The Fisher’s exact test or the Chi-square test was used to evaluate categorical data. The Mann–Whitney test or independent t-test was used to examine continuous data. Continuous data were presented as mean ± standard deviation. Univariate and multivariate logistic regression analyses were conducted to evaluate the effect of SFG + TtIE on fat necrosis adjusting for age, BMI, rate of fat grafting after TE-to-implant exchange (additional fat grafting after definitive implant placement), and follow-up.

Results

According to our inclusion criteria, 285 patients representing 474 reconstructions were initially included. One hundred and eighty-three reconstructions (38.6%) underwent SFG + TtIE, while 291 reconstructions (61.39%) were allocated in the control group (No-FGX). Before propensity score matching, the SFG + TtIE group had a lower proportion of reconstructions from patients with diabetes (p = 0.041), hypertension (p = 0.002), and patients receiving adjuvant chemotherapy (p = 0.001). The SFG + TtIE group had also a lower proportion of reconstructions for stage IIA, IIB, and IIIC breast cancer (p = 0.037); SSMs (p < .001); unilateral (p = 0.028), immediate (p = 0.014), subpectoral reconstructions (p < .001); and a reduced mean final TE volume (p < .001) (Table 1).

Table 1 Summary of baseline demographic and clinical characteristics before matching.
Full size table

After propensity score matching, 98 reconstructions that underwent SFG + TtIE were eligible for inclusion in this study and were 1:1 matched to reconstructions that did not undergo simultaneous fat grafting during TE-to-implant exchange (No-FGX group) (Table 2). Groups were satisfactorily matched for the demographic, oncologic, and surgical variables. Overall, the mean age was 51.3 ± 10.9 years, and most patients had a BMI lower than 30 kg/m2. The proportion of patients with past medical history of hypertension (23%), diabetes (5%), and active smoking status (4%) were not significantly different between groups. Most cases were performed after SSMs (84%) and were immediate (94%), bilateral (81%) reconstructions using a subpectoral plane (64%). Overall, 15% of reconstruction required postmastectomy radiotherapy, and 19% required adjuvant chemotherapy. Outcomes of the first stage of IBBR including the complications reported during the expansion process were comparable between groups (Table 3).

Table 2 Summary of baseline demographic and clinical characteristics after matching
Full size table
Table 3 Summary of clinical and surgical outcomes during the first stage of reconstruction after matching
Full size table

Intraoperative Outcomes of TE-to-Implant Exchange

After propensity score matching, the final TE volume achieved at the end of the first stage of reconstruction was comparable between groups (426.3 ± 139.6 cc versus 459.8 ± 149.2 cc, p = 0.223) (Table 3). Reconstructions in the SFG + TtIE group received larger implants during TE-to-implant exchange in comparison with the control group (539 ± 135.1 cc versus 495.97 ± 148 cc, p = 0.035), which was also supported by a higher mean implant–TE volume Δ in the SFG + TtIE group (112.7 ± 81.5 cc versus 36.17 ± 91.9 cc, p < .001). The total volume of fat injected during TE-to-implant exchange was 88.79 ± 41.51 ml in the SFG + TtIE group. The proportion of reconstructions that had different types of capsulotomies, partial capsulectomies, and total capsulectomies during TE-to-implant exchange (p = 0.33) and the proportion of reconstructions with smooth versus textured implants (p = 0.579) were not significantly different between groups after propensity score matching (Table 4).

Table 4 Summary of surgical outcomes of the second stage of reconstruction after matching
Full size table

Outcomes Following Exchange—Second Stage

After propensity score matching, a significantly higher proportion of reconstructions in the SFG + TtIE group underwent further fat grafting revisions following definitive implant placement when compared to the No-FGX group (19% versus 9%, p = 0.041). Nonetheless, the mean total volumes lipoinjected during revision procedures were not significantly different between groups (135.21 ± 90.5 ml versus 89.3 ± 35.2 ml, p = 0.19) (Table 4).

After propensity score matching, reconstructions from the SFG + TtIE group and the No-FGX group had comparable rates of postoperative hematoma (0% versus 0%), seroma (4% versus 1%, p = 0.369), implant infection (5% versus 4%, p = 0.497), capsular contracture (14% versus 6%, p = 0.057), implant malposition (7% versus 2%, p = 0.08), skin flap necrosis (1% versus 0%, p = 1.00), or dehiscence (4% versus 1%, p = 0.174) (Table 5).

Table 5 Summary of the complications of the second stage of reconstruction and oncologic after matching
Full size table

The rate of fat necrosis following TE-to-implant exchange was significantly higher in the SFG + TtIE group versus the No-FGX group (10% versus 2%, p = 0.017). The rates of breast cancer recurrence (3% versus 5%, p = 1.00) and breast cancer mortality (0% versus 3%, p = 0.496) were comparable between the SFG + TtIE group and No-FGX group. Overall, 4% of the IBBRs failed in the SFG + TtIE group while 3% of the IBBRs failed in the No-FGX group (p = 1.00) at a mean follow-up of 38.03 ± 22.62 months. The average follow-up was not significantly different between groups (35.84 ± 15.41 months versus 40.22 ± 27.96 months, p = 0.676) (Table 5).

On univariate analysis, SFG + TtIE was individually associated with an increased likelihood of fat necrosis (OR 5.45, 95%CI 1.16 to 25.58, p = 0.031). When incorporating the age, BMI, the rate of revisions with fat grafting procedures performed after definitive implant placement, and the follow-up into a multivariate regression model, SFG + TtIE persisted as an independent predictor for fat necrosis after definitive implant placement [second stage] (OR 6.55, 95%CI 1.23 to 34.95, p = 0.028) (Table 6).

Table 6 Predictors for fat necrosis after TE-to-implant exchange
Full size table

Discussion

Our matched cohort study for two-stage IBBR demonstrated comparable rates of complications following SFG + TtIE versus No-FGX, except for a significantly higher incidence of clinically notable fat necrosis after definitive implant placement in the SFG + TtIE group. To date, this is the largest study evaluating the surgical and clinical outcomes of simultaneous fat grafting during TE-to-implant exchange using a 1:1 propensity score-matching model which offers a consistent, accurate, and effective way to control possible confounder and effect modifiers for a homogenous pairwise comparison between groups. Furthermore, the comparable follow-up period between groups after propensity score matching also provides additional validity to our study results (35.84 ± 15.41 months versus 40.22 ± 27.96 months, p = 0.676).

Fat grafting as a single reconstructive modality offers an autologous, minimally morbid reconstructive approach; nonetheless, it usually requires several sessions to restore the shape and symmetry after mastectomy, even in small-volume breasts [4]. One of its major drawback is the unpredictable resorption rate [4]. On the other hand, when combined with alloplastic alternatives, it recreates natural esthetic outcomes and optimizes the implant’s interface with the mastectomy flaps and surrounding soft tissue [4]. Although previous reports have noted the implementation of serial sessions of autologous fat grafting during the expansion process and then during TE-to-implant exchange to optimize the outcomes of reconstruction [4], the most common practice to correct any contour deformity or to improve the mastectomy flaps is fat transfer after exchange or fat grafting simultaneously performed at the time of exchange (Fig. 2). Remarkably, although SFG + TtIE is an ingenious strategy to optimize the breast envelope and contour during definitive implant placement without increasing the risk of periprosthetic infection or wound-related complications, it may not provide definitive results (Fig. 3). In fact, our study demonstrated that after TE-to-implant exchange 6% of reconstructions required surgical interventions to address implant malposition, while 19% still required additional fat grafting procedures in the SFG + TtIE group.

Fig. 2
figure 2

A 45-year-old patient who underwent bilateral prepectoral two-stage IBBR. TEs were filled with 440cc of air at the end of the expansion process before TE-to-implant exchange (a frontal, b left oblique, c left lateral). Postoperative photographs 17 months following SFG + TtIE (right, 65 cc; left, 80 cc) with 450cc definitive implants (d frontal, e left oblique, f left lateral).

Full size image
Fig. 3
figure 3

A 66-year-old patient who underwent bilateral prepectoral two-stage IBBR. TEs were filled with 200cc at the end of the expansion process before TE-to-implant exchange (a frontal, b left oblique, c right lateral). Postoperative photographs 8 months after TE-to-implant exchange, without simultaneous fat grafting (No-FGX), and 335cc and 295cc definitive implants in the right and left breast, respectively (d frontal, e left oblique, f right lateral).

Full size image

We found that a higher rate of clinically relevant fat necrosis was reported in the SFG + TtIE group when compared with the No-FGX group even though similar fat grafting volumes were lipoinjected after exchange (135.21 ± 90.5 versus 89.3 ± 35.2, p = 0.191). In this setting, it is important to highlight that SFG + TtIE, individually (OR 5.45, p = 0.031) and during multivariate analysis (OR 6.55, p = 0.028), was an independent predictor significantly associated with increased odds of fat necrosis after definitive implant placement. Previous studies have also identified a higher incidence of palpable masses (38% versus 18.3%; p = 0.003) and requirement of postmastectomy imaging (47.3% versus 29%; p = 0.01) with fat grafting for breast reconstruction versus no fat grafting [11]. However, this did not translate into a significantly higher rate of biopsies performed for the histological characterization of fat necrosis or to rule out breast cancer recurrence (11.8% versus 7.5%; p = 0.32) [11].

Implant-based reconstruction of irradiated breast is associated with high postoperative capsular contracture rates [5, 12]. It is hypothesized that adipose-derived stem cells in lipofilling may play a major role in the therapeutic effect of autologous fat grafting against radiation-induced tissue changes [12]. Enhancement of periprosthetic tissue with fat grafting has shown promising results in reversing radiation-induced fibrosis via several mechanisms. These include decreasing collagen content (59% reduction), decreasing capsular density scores (p = 0.001), and decreasing fiber alignment scores (p <. 001) of implant capsules [13]. Nonetheless, in accordance with previous reports [12], no protective effect on the development of capsular contracture for two-stage IBBR was observed in our study. In this setting, further prospective, randomized studies are required to improve the assessment of the effect of autologous fat to reduce or prophylactically prevent capsular contracture in IBBR, especially taking into consideration that the proportion of adipose-derived stem cells in lipofilling may play a major role on the therapeutic effect of autologous fat grafting against radiation-induced tissue changes [12].

Interestingly, before propensity score matching a higher recurrence rate and mortality was reported in the No-FGX group when compared with the SFG + TtIE. The occurrence of this phenomenon was attributed to a significantly extended long-term follow-up, as well as a higher proportion of stage IIA, IIB, and IIIC breast cancers in the No-FGX when compared with the SFG + TtIE in the pre-matched analysis (Supplementary Material 1). After propensity score matching, the rates of breast cancer recurrence and mortality were comparable between groups.

Fat grafting does not induce neoplastic disease. The effect of progenitor adipose cells on small breast cancer foci remains of concern when fat lipotransfer is performed after the ablative treatment has been accomplished, especially in locally advanced tumors [14, 15]. Certainly, although this current report has a limited long-term follow-up, previous studies have shown comparable 5-year overall locoregional recurrence-free survival rates (95.1% [95%CI 89.6–100%] versus 96.2% [95%CI 92.1–100%]) and 5-year overall survival rates (94.6% [95%CI 88.7–100%] versus 95% [95%CI 89.6–100%]) between patients receiving fat grafting versus no fat grafting after breast reconstruction [11]. Still, prospective studies with an extensive follow-up are required [11].

Limitations

This study has a retrospective design for chart review and database synthesis. Due to the strict selection criteria and limited sample size, there is a latent probability for type II error. The impact of fat grafting during TE exchange, or as a revision procedure after definitive implant placement, was not evaluated with patient-reported outcomes measurements (PROMs). Volume retention and quality of the fat grafts were not assessed using histologic analysis or imaging. As patients were not randomized or blindly allocated into the intervention or control groups, there is a latent risk of selection or performance bias; however, this was minimized after propensity score matching. There are other factors that may have an impact on the incidence of complications in reconstructions performed with SFG + TtIE that we did not examine such as dose of postmastectomy radiotherapy, chemotherapeutic regimen, and inadvertent breast microtraumas causing fat necrosis.

Conclusion

SFG + TtIE is an alternative to improve the conus and envelope of reconstructed breasts during two-stage IBBR. SFG + TtIE does not increase the rate of periprosthetic infection or wound-related complication when compared with no fat grafting during TE-to-implant exchange. Additionally, it does not increase the rate of breast cancer recurrence at an average 3-year follow-up. Although SFG + TtIE is oncologically safe, a higher rate of clinically noticeable fat necrosis is expected, even if further fat grafting procedures are not performed after exchange.