Two-stage Implant-based Breast Reconstruction Without the Use of Tissue Expanders

INTRODUCTION

Although the use of TEs followed by a definitive implant is the most common approach for breast reconstruction after mastectomy,¹ their use has many disadvantages, including increased cost,² patient discomfort, and inconvenience, with the requirement for multiple office visits for filling with increased risk of complications including tissue expander (TE) rupture and infection.^2,3 Furthermore, increased rates of reconstructive failure in patients requiring postmastectomy radiotherapy have been documented,^4,5 as well as chest wall deformity⁶ and, rarely, extrusion through the skin.³ Direct-to-implant reconstruction (DTIR) obviates many of these drawbacks but can arguably lead to increased rates of complications and reconstructive failure, especially in higher risk patients, when immediately placing a definitive implant under tenuous mastectomy flaps.^7–11

For many patients, the use of a TE has nothing to do with tissue expansion, as they have undergone a skin-sparing mastectomy (SSM) or nipple-sparing mastectomy (NSM) that could accommodate a definitive implant.^2,3 Surgeons use TEs to minimize the risk of ischemic complications by reducing tension on the mastectomy flaps and to create a stable mastectomy pocket that can be modified in a second stage with fat transfer, capsulorrhaphy, capsulotomy, and definitive implant placement.³ In our practice, since 2016, 2-stage implant reconstruction using a TE has been reserved for high-risk patients who might not be good candidates for DTIR.¹² In 2020, we modified our practice, reasoning that first-stage placement of a low-projection, appropriate base-width silicone implant might achieve the same benefits (decreased ischemic stress) as a TE, with decreased cost, increased patient comfort, and convenience without the requirement for multiple office visits for TE filling. All patients subsequently underwent second-stage implant exchange to achieve their goal reconstruction size (Figs. 1–4). Here, we report on 150 consecutive patients where this implant-only (IO) approach was utilized to provide patients with a safe, 2-stage prepectoral reconstruction after mastectomy without the use of TEs.

PATIENTS AND METHODS

The author retrospectively reviewed patients in his practice who underwent SSM or NSM and immediate 2-stage (either a TE followed by a definitive implant, or a smaller definitive implant followed by a planned second surgery for a larger implant) prepectoral implant-based breast reconstructions between January 2016 and January 2024, excluding patients who underwent delayed reconstructions or single-stage DTIRs, with a minimum of 6 months follow-up. All procedures, including both oncological resection and reconstruction, were performed by the author. The IO approach was used for patients who would have previously undergone first-stage TE reconstruction in the author’s practice and were not good candidates for DTIR (body mass index [BMI] > 35 kg/m², previous radiotherapy, mastectomy weight greater than 600 g, grade 3 ptosis, or the requirement for a definitive implant larger than 600 mL). Active smokers and patients with uncontrolled diabetes (HgBA1c > 7) were not offered immediate reconstruction. In some situations, an intraoperative decision was made based on the unexpected requirement for a larger definitive implant, concern regarding mastectomy flap perfusion, or a heavier-than-predicted mastectomy weight. The following patient variables, operative, and oncological details were recorded: age, BMI, patient comorbidities (diabetes, hypertension), therapeutic versus prophylactic mastectomy, history of breast radiotherapy, mastectomy specimen weight, ptosis grade, SSM versus NSM, the incision used (inframammary, Wise-pattern, lateral radial, or transverse elliptical), final implant volume, the use of acellular dermal matrix (ADM), synthetic mesh or no mesh, and the type of lymph node dissection performed. The administration of neoadjuvant or adjuvant chemotherapy and postmastectomy radiotherapy was also recorded. Complications were separately recorded for first- and second-stage implant procedures. Complications were recorded as either minor (requiring only outpatient care) or major (requiring hospital admission, operative intervention, or intravenous antibiotics) and included hematomas, seromas, infections, wounds (including dehiscence and mastectomy flap necrosis), and reconstructive failure (requiring implant removal with or without replacement). We then compared patient demographics, surgical factors, oncological treatments, and surgical complications between 2-stage reconstruction using the traditional first-stage TE approach versus our IO approach, described here using definitive implants in both stages. We then evaluated the risk of reconstructive failure and its association with patient demographics, comorbidities, surgical factors, oncological treatments, and the use of the IO and TE approaches. Statistical analyses were performed using IBM SPSS version 25.0 (IBM Corp., Armonk, NY). One-tailed t tests were used for continuous variables, and the Fischer exact or chi-square tests were used for categorical variables, with a P value of less than 0.05 considered significant. Univariate analysis was performed to determine which risk factors were significant for reconstructive failure. These factors, along with IO and TE variables, mastectomy specimen weight, and BMI were used to perform a stepwise multivariate logistic regression analysis to determine whether the IO protocol described here was an independent predictor of reconstructive failure (or, conversely, could safely replace the TE protocol).

SURGICAL TECHNIQUE

All patients underwent NSM using an inframammary, lateral radial incision, or a Wise-pattern mastopexy.¹³ In most cases, either a 16 × 20 cm, medium thickness ADM (AlloMend, reference no. 77583320 [Allosource, Centennial, CO]), Alloderm (Select Tissue Matrix, reference no. 1518320P [LifeCell Corporation, Branchburg, NJ]), or synthetic mesh (TIGR Matrix, reference no. NSTM1520 [Novus Scientific, Uppsala, Sweden]) was utilized to stabilize and define the implant pocket. The ADM or mesh was wrapped around the TE or silicone implant ex vivo before placement above the pectoralis muscle. No TEs were used after March 2020 for reconstruction after SSM or NSM when we eliminated their use for 2-stage reconstructions. After 2020, all patients who underwent planned 2-stage reconstruction had definitive implants placed at both stages, unless they had a skin deficit requiring TE. Intraoperatively, for patients who were determined to be poor candidates for immediate reconstruction, a delayed-immediate¹⁴ or delayed reconstruction was performed, and these patients were excluded from the study. Lateral adipose tissue was sutured down to the chest wall to reinforce the lateral mammary fold and minimize the risk of seroma, which was also especially important when ADM or mesh was unavailable. Sizers were used to determine the appropriate base width and then the corresponding Natrelle INSPIRA (AbbVie, Chicago, IL) smooth, responsive, low-projection, or Mentor (Johnson and Johnson MedTech, Irvine, CA) moderate classic implant was placed into the prepectoral plane. Two 15-round drains were placed. Patients were placed in an adhesive bra and were discharged on the day of surgery. Drains were removed when output was less than 30 mL with all drains removed 3 weeks postoperative regardless of output. Postoperative antibiotics were continued for 24 hours after surgery (oral antibiotics upon discharge). Implant exchange occurred between 4 weeks and 6 months after the surgery for smooth, round, moderate, full, or extra-projection implants.

RESULTS

We identified 305 consecutive patients who underwent 457 immediate 2-stage prepectoral breast reconstructions over 6 years. There were 155 patients who underwent conventional 2-stage reconstruction with a first-stage TE and 150 patients who underwent the modified IO approach described here. The average patient age (52.6 ± 8.5 versus 50.9 ± 10.2 y, P = 0.255), BMI (30.5 ± 6.3 versus 32.2 ± 8.5 kg/m², P = 0.441), proportion undergoing unilateral versus bilateral reconstruction (0.19 versus 0.23, P = 0.575), mastectomy weight (685 ± 266 versus 620 ± 306 g, P = 0.555), and final implant size (602 ± 127 versus 649 ± 107 mL, P = 0.354) were similar between the conventional first-stage TE and IO groups, respectively (Table 1). There was a greater percentage of patients in the TE group who had a formal diagnosis of diabetes (21% versus 11%, P = 0.017) and hypertension (54% versus 37%, P = 0.003). There was a significantly higher percentage of patients in the IO group who had synthetic mesh placed (81.5% versus 13.5%, P < 0.00001), whereas the TE group had a higher percentage of ADM placed (65.2% versus 9.3%, P = 0.001) and patients who underwent reconstruction with neither implant nor synthetic mesh (21.3% versus 9.2%, P = 0.04). The choice of surgical incision was similar between the groups (Table 1) with the inframammary fold incision being the most common for both groups. The proportion of patients undergoing neoadjuvant and adjuvant chemotherapy and postmastectomy radiotherapy was similar between the 2 patient cohorts (Table 1). Patients in the IO cohort more frequently underwent no axillary surgery and fewer axillary dissections than the TE cohort (Table 1). The average follow-up time was shorter in the IO cohort (16.7 ± 6.4 versus 26.2 ± 9.1 mo) than in the TE cohort (P = 0.02). The ratios of patients undergoing therapeutic surgery were higher in the IO cohort (Table 1).

COMPLICATIONS

A total of 146 (48%) experienced a complication, and 19 patients (6%) experienced at least 2 major complications (Table 2). The complication rates for the TE (84 [54%] patients) and IO groups (71 [47%] patients) were similar (P = 0.231) (Table 2). Seventy-three (50%) patients with complications required reoperation, inpatient admission, or intravenous antibiotics (major complications), of which 47 patients (64%) required implant removal or replacement, for an overall reconstructive failure rate of 15% (Table 2). The rates of major and minor complications for both the TE and IO cohorts are presented in Table 2 with no significant differences noted between the 2 groups, except for a higher rate of minor postoperative wounds in the IO group and a higher rate of seromas requiring operative drainage in the TE group. The incidence of complications of major and minor hematoma, minor seroma, major and minor infections, and major postoperative wounds was similar between the IO and TE cohorts. The requirement for implant removal or reconstructive failure (with or without successful replacement) was similar between the TE and IO groups. The overwhelming majority of complications in both groups occurred after the first stage. In the IO group, 6 (10%) complications occurred after the second stage (3 minor wound healing complications, 2 minor infections, and 1 major infection requiring implant removal). In the TE group, there were 4 (6%) minor wound healing complications after the second stage. There was no significant difference between patients in the TE and IO groups, 145 (94%) and 147 (98%), respectively, who achieved successful implant-based reconstruction (P = 0.0686).

The incidence of reconstructive failure (requiring implant removal or replacement) was significantly greater in patients with higher BMI (33.8 versus 30.6 kg/m²; P < 0.004) and those with higher mastectomy weights (756 versus 617 g; P < 0.003). Furthermore, reconstructive failure was more common in those who underwent Wise-pattern mastectomy, axillary dissection, therapeutic surgery, the use of ADM, or postmastectomy radiotherapy (Table 3). Patients with diabetes, grade 3 ptosis, and a history of prior breast radiotherapy were also at a higher risk of reconstructive failure. There was no statistically significant difference between those patients who experienced reconstructive failure and successful reconstruction with regard to age, final greater implant volumes, hypertension, NSM or SSM, or unilateral versus bilateral surgery. Although the use of ADM was a risk factor for reconstructive failure, the use of synthetic mesh was not. Patients who underwent neoadjuvant or adjuvant chemotherapy were also not at a higher risk of reconstructive failure. The use of the IO or TE approach did not impact the risk of reconstructive failure on either univariate or multivariate analysis (Tables 3, 4).

DISCUSSION

There have been several studies demonstrating comparable outcomes and patient satisfaction between DTIR and 2-stage reconstruction with TEs.^15–17 Despite this, most reconstructive surgeons still use a 2-stage approach to implant-based breast reconstruction using a first-stage TE.¹ Although patients with a skin deficit require tissue expansion or skin replacement,¹⁸ most patients who undergo an SSM or NSM are candidates for either approach. Most surgeons use the 2-stage approach to reduce ischemic stress on the mastectomy flaps, to allow for more patient input regarding final implant choice, and to provide a stable pocket for modification with capsulotomy and/or capsulorrhaphy, fat transfer, and implant exchange.^2,3

Previous studies have demonstrated that DTIR may have a higher risk of complications in patients with higher BMIs and larger breast sizes, those with a history of or requiring radiotherapy, and those requiring larger implant sizes.^7–12 In the author’s practice, patients with BMIs greater than 35 kg/m², history of radiotherapy, grade 3 ptosis, mastectomy specimen weights greater than 600 g, or requiring implants greater than 600 mL were historically offered 2-stage reconstructions with a first-stage TE to mitigate the risk of complications. In May 2020, we replaced first-stage TEs with definitive silicone implants of appropriate base width (or slightly narrower) but lower projection than the final desired implant size. This approach was intended to provide the benefits of TEs (reduced ischemic stress, a stable mastectomy pocket for second-stage modifications, and increased patient input on final reconstructed breast size) while also eliminating the drawbacks (increased cost, inconvenience for patients requiring regular office visits for TE filling, discomfort, risk for infection, and deflation). In our practice, TEs are 3 times the cost of fourth-generation silicone implants. Although one could argue that TEs offload the mastectomy flaps to a greater degree than our approach, we demonstrate a similar risk of complications in our practice in high-risk patients undergoing staged reconstruction with a TE versus the 2-staged IO approach presented here. Most empty silicone TEs have a weight of 40–60 g and are typically filled at the time of surgery with variable amounts of saline.^2,19,20 In the standing position, these partially filled expanders place most of the ischemic stress on the lower pole of the reconstructed breast, where our incisions lie (95% of our mastectomy incisions are either in the inframammary fold or are Wise patterns). Definitive silicone implants have their weight more equally distributed between the upper and lower poles of the breast, placing less stress on an inframammary incision. Admittedly, TEs can be filled with air instead of saline, which would minimize the preferential stress on the lower pole but has shown conflicting results with regard to reducing the risk of complications compared with TE filling with saline.^21–23 Our approach also allows patients to delay their second-stage surgery indefinitely, as there is no strict requirement for implant exchange, whereas TEs are recommended to be replaced within a certain time frame (Figs. 5, 6). Several studies have also documented a lower risk of reconstructive failure for DTIR versus the 2-stage approach in patients undergoing postmastectomy radiotherapy.^4,5 Our approach provides patients who require postmastectomy radiotherapy an option that both reduces their risk of reconstructive failure in comparison to a 2-stage TE reconstruction and decreases ischemic stress on their mastectomy flaps in comparison to a more standard DTIR. Our report demonstrates an elevated risk of complications of nearly 50% in both the IO and TE groups. Although this is a higher complication rate than seen in an average-risk patient cohort,^7–9 this complication rate is consistent with the high-risk nature of this patient population, which has an elevated risk of both complications and reconstructive failure.¹² This may also be secondary to the usage of more Wise-pattern and inframammary incisions in obese patients with macromastia, ptosis, and longer mastectomy flaps. Regardless, our reconstructive failure rate of ~15% is certainly within the acceptable range for this high-risk patient cohort.¹²

Although we selectively use the approach described here in high-risk patients who have previously undergone TE reconstruction followed by implant exchange in our practice, many surgeons might choose to use this approach if they prefer 2-stage TE reconstructions for all patients. There are still potential benefits of the use of TEs over the approach described here, including the ability to completely deflate the prosthetic in the event of ischemic complications and the even lower ischemic stress that an air-filled TE places on tenuous mastectomy flaps. In addition, TEs also have tabs for fixation, which are not available for definitive implants, reducing the need for meshes and ADM often required in DTIR.^2,19,24 This may significantly reduce the cost of the TE approach and potentially make the IO approach more expensive if mesh is required. Although we prefer to use mesh with the IO approach, we are often restricted from using it in our ambulatory surgery center. In these cases, we advance soft tissue to reinforce the lateral mammary fold and use a postoperative tape bra to reduce the risk of implant malposition. Regardless, we feel that most of our patients are sufficiently offloaded by using low-projection implants in the first stage with similar rates of complications to those we experienced in a similar high-risk group in our practice that previously underwent 2-stage breast reconstruction using TEs.

This report has several limitations, as it was a single-surgeon, single-institution experience where the author performs both the mastectomy and reconstruction, which is unusual, and our findings may not be applicable in other practice settings. Furthermore, our definition of “high-risk” patients was arbitrary, and there may be a group of patients who preferentially benefit from TE placement that we failed to identify in this study. In addition, although the TE and IO groups were, for the most part, well matched, there was significantly more use of biosynthetic mesh in the IO group as our practice evolved to less routine use of ADM. The use of ADM is known to increase the risk of complications in implant-based reconstruction, including reconstructive failure,^25,26 as we have also demonstrated here, which resulted in a warning by the Food and Drug Administration.²⁷ To confirm that the IO approach had similar risks of reconstructive failure as the conventional TE approach, and that the reduced use of ADM in the IO group did not confound our results, we performed a multivariate analysis, which confirmed that both the IO and TE variables did not significantly contribute to the risk of reconstructive failure (Table 4).

CONCLUSIONS

Although the most common approach to postmastectomy reconstruction requires 2 stages involving a first-stage TE, we have shown here that a first-stage low-projection definitive silicone implant can safely replace the TE, even in women at high risk for complications. This affords the women the safety and benefits of a 2-stage reconstruction² without the drawbacks of a TE, which include increased cost and patient discomfort²; the inconvenience and potential complications associated with repeated office visits for filling; the requirement for timely scheduling of the second-stage implant exchange; the possibility of significant chest wall deformity⁶; and the increased rate of complications associated with radiating TE reconstructions.^4,5

DISCLOSURE

The author has no financial interest to declare in relation to the content of this article.

REFERENCES