Introduction

Implant-based reconstruction remains the most common immediate breast reconstruction technique [1].

Traditionally, breast reconstruction techniques have been comprised of a two-stage procedure: a tissue expander (TE) insertion at the time of mastectomy and after full expansion, usually a few months later, a second procedure to exchange the expander to implant. However, over the years there has been increased interest in direct-to-implant (DTI) procedures, namely inserting the final implant immediately at the time of the mastectomy [2].

The appeal of DTI procedure is obvious: immediate full breast shape, elimination of the need for multiple visits required for expansions as well as elimination of a second operation for the exchange of the tissue expander to implant as in a traditional two-stage expander/implant reconstruction. As such, DTI may inherently reduce costs [3]. However, there is much variety in previously published results regarding the complication rate of DTI as compared to [4,5,6,7,8]. In addition, the lack of familiarity with the procedure and hesitancy of putting greater tension on the mastectomy skin flaps may have prevented many surgeons from using this newer method.

The conflicting results regarding experience with DTI can be explained by the variability of a number of factors, including: patient population, variability in surgeons with varying experience and different techniques (pre-pectoral versus sub-pectoral), materials used (different kinds of implants and artificial dermal matric (ADM)) and steep learning curve required for DTI.

The goal of this study was to focus on the re-operation rate of DTI in comparison with two-stage TE reconstruction and trying to eliminate confounders by including only cases performed under the supervision of the same senior general and plastic surgeons, using the same plain for implant placement (supectoral) and the same sort of implant and ADM.

Methods

This retrospective, observational cohort study evaluated consecutive 165 patients (222 breasts) who had skin sparing mastectomy due to breast cancer or risk reducing mastectomy with or without nipple areola complex preservation and immediate implant-based reconstruction at Hadassah Medical Center by the same supervising senior general surgeon and plastic surgeon (EC, NA) between January 2010 and December 2019. Patients included were only those who had follow-up of at least 2 years and sub-pectoral implant or expander placement with ADM (ADM—AlloDermTM, textured implants/ smooth round tissue expanders—MENTOR®). Patients were divided into two groups according to the type of reconstruction done: 42 breasts underwent two-stage tissue expander procedure, and 180 underwent single-stage direct-to-implant procedure.

The selection of the type of reconstruction was based on clinical judgement and preference of the reconstructive surgeon. Pre-operative selection to reconstruct with TE included patients with higher risk for mastectomy flap vascularization problems such as smokers and also patients with the cancer involving the skin and thus the need to remove skin. During the operation, mastectomy flaps were evaluated clinically by the reconstructive surgeon. In case of bluish discoloration of the skin or extended areas with very thin skin with dermis only, an expander was chosen over DTI. The size or shape of the breast was not part of decision process regarding TE versus DTI. Breast ptosis and volume were taken into consideration when to offering skin sparing only versus nipple areolar complex sparing. Women with larger ptotic breasts in our practice are offered either skin sparing mastectomy or if safe oncologicaly to delay the operation than a two or three stage operation is planned: first bilateral reduction and then 3-4 months later mastectomy with TE or DTI. This last group of patients was excluded from this study.

After receiving permission from the hospital ethical committee (HMO-0065-22) patients medical records were reviewed and data were collected including: data from post-operative clinic follow-up, surgery reports, emergency room visits, oncological clinic visits and any other visit. Since the standard follow-up from mastectomy and reconstruction was 2 years, this was set as the follow-up timeframe. Specific data collected included patient demographics such as age, BMI, smoking, diabetes, the presence of BRCA gene mutation, mastectomy type, extent of the lymph node intervention and chemotherapy or radiotherapy the patient received.

Outcomes data collected at 2-year follow-up included complications classified by major or minor complication and classification according to Clavien–Dindo, capsular contracture Baker 3-4, re-operation due to complication or due to patient’s wish to improve aesthetic appearance. Re-operation in case of TE was defined as further operation after the second stage of TE exchange to implant. Minor complications were defined as complications that did not require hospitalization or invasive procedure, while major complications did. Minor complications included hematoma or seroma not requiring surgery or drainage, infection or skin exposure not requiring surgery or hospital admission. Major complications included expanding hematoma, seroma requiring drainage, infection requiring hospital admission or surgery, skin necrosis, implant exposure and implant failure. Re-operation due to complication included also reoperation due to implant rapture and capsular contracture. Since there were a small number of patients with unilateral reconstruction that chose to operate on the contralateral breast for symmetry and all of them had additional procedure on the reconstructed breast at the same time to improve aesthetic results, we therefore included them in the re-operation group due to aesthetic reasons.

Statistical Analysis

Statistical analysis was performed using SPSS program, version 25.0.

Revisional surgery rate was calculated with a 95% CI. To test the association between two categorical variables, the Chi-square test as well as the Fishers exact test was applied. The comparison of a quantitative variable between two independent groups was made using the T-test. Variables, which were found to be significantly associated with a dichotomous dependent variable (re-operation, complications) in the univariate approach, were entered into a multivariate logistic regression model.

Results

Of 222 breasts operated, 73 breast reconstructions overall were prophylactic mastectomies, while 149 were due to breast carcinoma. Mean patient age was 46.7 (range 25–76 years), and mean body mass index was 25.5 kg/m2 (SD +/− 4.69) with no significant difference between the groups. Forty patients (18%) received post-operative radiotherapy with no significant difference between the groups. There were significantly more prophylactic mastectomies and BRCA gene mutation in the direct-to-implant breast reconstruction group, and more active smokers and diabetic patients in the tissue expander group (Table 1).

Table 1 Patients’ characteristics and operative data overall and by group
Full size table

Most procedures (161, 73.2%) were skin sparing mastectomies. In the direct-to-implant breast reconstruction group, 130 procedures (72.2%) were skin sparing, and 42 (23.3%) were nipple sparing. In the tissue expander/implant group, 31 (77.5%) were skin sparing and 7 (17.5%) were nipple sparing (p=0.714). There was no significant difference in the complication rate between the groups, 47 (26.1%) for DTI and 17 (40.5%) in the TE group; however, major complication and re-operation rate due to complications were significantly different, 30 (16.7%) and 19 10.6% for DTI and 16 ( 31%) and 11 (26/2) for the TE group, respectively (p= 0.035, p = 0.008) (Table 2).

Table 2 Post-operative complications and re-operations
Full size table

No significant differences were found in the Clavien–Dindo stages 1-3a between the groups, meaning complications requiring non-surgical treatment such as antibiotics or complications treated under local anaesthesia such as drainage of seroma. Only grade 3b, which is defined as complications requiring treatment under general anaesthesia, was found to be significantly higher at the TE group (19 (10.6%) for DTI and 11 (26.2%) for TE Group, p=0.008). This is similar to our definition of re-operation rate due to complication.

There was strong correlation (OR=3.38) between re-operation due to complication and adjuvant chemotherapy (p=0.045). Correlation between re-operation and other variables including smoking, diabetes, post-operative radiotherapy and BMI was not found to be significant.

Re-operation for aesthetic improvement was significantly higher in the TE group-69 (38%) in DTI and 29 (69%) in TE group, p=0.0003). The mean number of re-operations per breast for aesthetic dissatisfaction in each group, not due to direct complication or capsular contraction, was 0.64 (SD +/− 1.08) in the DTI group and 1.1 (SD +/− 1.078) for TE which was significantly higher (p=0.016).

Further operations for improving aesthetic appearance included fat graft, augmentation, liposuction, excess skin removal, exchanging the implant, scar revision and nipple reconstruction (Table 3, Figs. 1 and 2). Additionally, we included breast reduction and mastopexy to the opposite side in unilateral cases (only 5 patients in the DTI group and 2 patients at the TE group). Additional two patients who wished to have their implants removed completely and remain without reconstruction or exchanging the reconstruction to an autologous reconstruction were also included.

Table 3 Distribution of procedures to improve overall aesthetic appearance
Full size table
Fig. 1
figure 1

A 45 y/o, before (left photograph) and 6 months s/p (right photograph) prophylactic mastectomy due to BRCA mutation and reconstruction with anatomic implant (MENTOR® 390cc, CPG 313) and AlloDermTM . This patient was pleased with the result and did not want any further operations

Full size image
Fig. 2
figure 2

A 52 y/o, s/p right mastectomy and reconstruction with anatomic implant (MENTOR® 295cc, CPG 322) and AlloDermTM. 6 months later she had nipple reconstruction (left photograph, 18 months after first operation) and a year later scar revision and fat injection to the right side and left mastopexy (right photograph, 3 months after second operation). The patient was still not completely satisfied with the symmetry of breasts; however, she decided not to go into further operations

Full size image

Discussion

One potential benefit of DTI in comparison with the traditional two-stage TE/implant reconstruction is its being a single-stage operation, avoiding the multiple visits for expansion and a second surgery for TE exchange. Our study shows significantly lower major complications and fewer re-operations due to complications in DTI (16.7%, 10.6%, respectively) compared to TE (31%, 26.2%, respectively). Our low re-operation rate of 10.6% following a two-year follow-up can reassure patients as well as reconstructive surgeons that DTI largely remains a single-stage operation in regard to unavoidable additional operations. However, the rate of re-operation indicated only by the patient’s desire for improved aesthetic result is quite high in DTI, with more than one-third of the patients (38.3%) wishing additional invasive procedures for better results. Despite this, it remains significantly lower than the two-stage technique (69%) with an average of 0.64 additional surgeries per patient in the DTI group and 1.1 in the TE group.

The findings regarding complications and re-operation in both groups vary in the literature. A meta-analysis by Basta [4] et al. including 13 studies that compared DTI to two-stage reconstruction found that DTI had higher risk for flap necrosis and twofold risk for implant loss. However, recent comparative studies from tertiary centres showed the same number of re-operations and less or the same rate of complications for DTI [7,8,9] . Furthermore, there was no difference in patient satisfaction [5, 6, 10].

This variability in the published literature can be at least partially explained by the learning curve needed for DTI. Results from specific centres specializing in breast that have already gained enough experience and have better patient selection may show lower complications rate.

Although we tried to eliminate different operative variables in this study, a selection bias remains, as reflected in the higher percentage of smokers and diabetics in the two-stage group and more prophylactic cases in the DTI group. There was no difference in post-mastectomy radiation therapy between the groups. This can be explained by the fact that radiation was not part of the considerations for selection of reconstructive technique. Our experience with DTI is that it is a viable option when post-operative radiation is expected, especially with the use of ADM which provides some protection from post-radiation capsular contracture.

Interestingly, in logistic regression analysis only adjuvant chemotherapy was correlated with re-operation due to complications. Perhaps this can be explained by the careful patient selection so that those with increased pre-operative risk for complications, such as smokers and diabetics underwent a two-stage reconstruction, thereby reducing the negative effect of those factors. For example, smoking is known to reduce the vascularity of the mastectomy flap. As such, gradually increasing tension on the flaps via TE may help to avoid these negative effects of smoking. However, adjuvant chemotherapy, which may affect the immune system and infection rate, is not a factor in selection of technique, as it may affect either equally, and does not appear to proffer added benefit of one technique over another.

Another major consideration that may have affected patient selection is the intraoperative evaluation of the mastectomy flaps by the plastic surgeon immediately after the mastectomy. In our practice, only well-vascularized flaps were selected for DTI. The selection therefore was made based on the plastic surgeon’s judgement, without assistance from imaging devices such as SPY (Novadaq Technologies, Inc., Bonita Springs, Fla.). The above-described selection bias only places further emphasis that the results are heavily determined by the experience of the surgeon and decision of patient selection. In properly selected cases, DTI can be a safe technique with 90% of the patients requiring no re-operation due to complications.

A weakness of this study is the relatively short follow-up of only 2 years. Longer follow-up may reveal higher percentage of re-operation due to implant rapture and capsular contracture. However, it can be assumed that long-term implant rapture and capsular contracture are not expected to be different between the groups.

Conclusion

Direct-to-implant immediate breast reconstruction can provide a good alternative to the traditional two-stage TE/implant operation in experienced dedicated breast centres. Both patients and surgeons can be reassured that re-operation rate due to complications or aesthetic improvement is lower in DTI, and in the majority of the cases, DTI is indeed one-stage reconstruction.