Background

The aim of a mastectomy is the complete removal of fibroglandular breast tissue. Yet, due to the frequent irregularity or lack of the superficial fascia of the breast – which is considered as the landmark of dissection – fibroglandular breast tissue remains [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18]. Furthermore, in some cases the subcutaneous tissue layer is simply too thin to allow total removal of breast tissue and simultaneously preservation of a well-perfused skin flap. Papassotiropoulos B et al. [12] demonstrated a distance of below 1 mm between the mastectomy specimen surface and breast tissue to be significantly associated with higher residual fibroglandular breast tissue (RFGT) rates in the skin flap.

RFGT has been observed in up to 100% of studied patients [19] and to amount up to 26% of the preoperative fibroglandular tissue (FGT) [3]. Unifocal [4] as well as multifocal dissemination [20] of RFGT has been described with various predilection sites including the craniolateral quadrant [3], the caudolateral quadrant [18] as well as the retroareolar region [19].

RFGT worsens the prognosis of the patient as it is associated with the remaining of occult breast cancer at the time of mastectomy as well as an increased local recurrence risk indefinitely after the surgery [7, 10, 16, 17, 21, 22].

In view of the oncologic consequences of RFGT and the lack of international consensus on the prevention of it – identification of patients at risk for RFGT in a first step is of uttermost clinical importance. Until now, only few authors have evaluated risk-factors for RFGT – and reported contradictory results [1,2,3,4, 8, 10,11,12, 14, 17,18,19,20, 23, 24]. Inconclusive findings were – among other parameters – described for type of mastectomy – which is in view of the rising nipple (NSM)- and skin-sparing mastectomy (SSM) rates particularly worth further studying.

Notably, besides the incomplete removal of FGT in the course of a mastectomy – the risk of a too radical surgical approach with additional resection of the subcutaneous tissue and subsequently increased risk of ischemic complications has been implied in literature and needs further evaluation [22].

Aims

The primary aim of the study was to assess a potential difference in RFGT volume between the different types of mastectomies (radical mastectomy (RME) vs. SSM vs. NSM). The secondary objectives of the study were to identify other potential risk factors for RFGT following a mastectomy and to assess the postoperative skin envelope thickness in comparison to preoperative measurements.

Methods

We performed a systematic literature search in PubMed using the MESH terms [residual fibroglandular breast tissue], [residual breast tissue], [mastectomy] and [risk factor] to provide an overview of risk factors for RFGT following a mastectomy.

We furthermore, performed a retrospective analysis of patients with a therapeutic or prophylactic mastectomy that were treated at the Department of Obstetrics and Gynecology of the Medical University of Vienna, Austria, between 01.01.2015 and 26.02.2020 and had an archived postoperative breast magnetic resonance imaging (MRI) examination. Patients with a second-look resection, an autologous breast reconstruction prior to the postoperative breast MRI and patients without an available post-mastectomy MRI or clinical data were excluded from the retrospective analysis.

Breast MRI examinations were done on 1.5T or 3T scanner following international guidelines [25], with dedicated coils and patients lying in a prone position. Measurements were conducted by an experienced breast radiologist. Pre-contrast T1-weighted 3D gradient-echo sequences and T2-weighted Turbo-Spin echo sequences were used to quantify RFGT. In addition, when available, preoperative MRI was used to aid in the correct identification of RFGT. MRI measurements included preoperative whole breast (including the skin envelope and nipple-areola-complex (NAC)), preoperative FGT, postoperative RFGT volume and the pre- and postoperative tissue envelope thickness.

Volumes were calculated with the help of a semi-automated segmentation with a dedicated software (ITK-SNAP) [26]. Measures are reported in mm3. RFGT thickness was determined in the retroareolar region and in the medial and lateral aspect of the breast. Measures are stated in mm.

Mean, standard deviation, minimum, maximum, median and interquartile range were used to describe metric variables. Frequencies and percentages were evaluated to describe nominal scaled parameters. The correlation between two nominal scaled variables was tested using Chi-square test – and if required - Fisher’s exact test. To test the difference between metric variables between 2 groups the students T-test and in case of skewed distribution of data Mann-Whitney-U-test were applied. In case of more than 2 groups the Kruskal-Wallis test was used.

Methods were previously presented in more detail in a study on the oncologic implications of RFGT based on the same study population [22].

The study was approved by the ethics´ committee of the Medical University of Vienna (Ethikkommission Medizinische Universitaet Wien, EK-Nr. 1067/2020). The need for informed consent was waived due to the retrospective nature of the study. The study was performed in accordance with the Declaration of Helsinki.

Results

135 patients (181 breasts) of 737 screened patients (897 breasts) met the inclusion criteria and were enrolled in the study. An overview of excluded patients is presented in Supplement Materials (see Table S1).

A RME was performed on 81 breasts (44.8%), a NSM on 66 breasts (36.5%) and a SSM on 34 (18.8%) breasts. Median follow-up duration in the RME subgroup was 47.3 months (IQR 26.4; 82.4), 26.1 months (IQR 11.8; 44.2) in the SSM and 25.2 months (IQR 18.5; 38.0) in the NSM subgroup. Patient characteristics are displayed in Table 1. Details regarding the surgical procedure are presented in Table 2.

Table 1 Patient characteristics
Full size table
Table 2 Surgery
Full size table

91.4% (74/81) of RMEs were performed with an oncologic indication, compared to 61.8% (21/ 34) of SSMs and 47.0% (31/66) of NSMs. Details regarding breast cancer characteristics are displayed in Table 3.

Table 3 Breast cancer characteristics
Full size table

RFGT was present in 93.9% of breasts following a NSM, 91.2% of breasts after a SSM and 85.2% of breasts following a RME.

The median time between the preoperative and postoperative MRI was 15.7 months (range 12.6; 27.0) in the RME subgroup, 13.5 months (range 12.4; 15.4) in the SSM cohort and 14.0 months (range 11.9; 22.5) in the NSM subgroup (p = .093). Details regarding MRI measurements are outlined in Table 4.

Table 4 MRI measurements
Full size table

Regarding the volume of RFGT – statistically higher volumes were present in the NSM-cohort (1414.0 mm3 (230.0; 3668.0)) compared to the RME (100.0mm3 (820.0; 1142.0)) and SSM-subgroup (167.0mm3 (20.0; 928.0)) (p < .001).

Furthermore, the following parameters were significantly associated with RFGT volume in univariable analysis: reconstruction (p = .012) – with higher RFGT volumes in breasts with immediate primary fixed volume implant reconstruction compared to immediate primary tissue expander, delayed primary tissue expander and no reconstruction; ADM or mesh (p = .031) – with higher RFGT volumes in breasts reconstructed with an ADM compared to synthetic mesh and no ADM or synthetic mesh; age (p = .022) – with higher RFGT volumes in younger patients; FGT volume (p = .012) – with higher RFGT volumes in case of higher FGT volume and whole breast volume (p = .030) – with higher RFGT volume in case of higher whole breast volume.

No significant association with RFGT volume in univariable analysis was found for: ME (mastectomy) indication (oncologic vs. prophylactic; p = .161), ME volume (p = .935), lymph node surgery (sentinel node vs. axillary dissection vs. no lymph node surgery; p = .409), surgeon (p = .214), BMI (p = .507), prior breast operation (breast conserving surgery vs. mastopexy/reduction mammaplasty vs. breast implants vs. no prior breast operation; p = .104), neoadjuvant chemotherapy (p = .636), prior ME radiotherapy (p = .477), post ME radiotherapy (p = .848), extent of disease (monofocal vs. bifocal vs. multifocal vs. multicentric; p = .809) and T-stage (p = .611).

The observed reduction in the postoperative compared to preoperative skin and subcutaneous fat tissue thickness measured medially and laterally reached statistical significance in the NSM cohort (medial 9.5 mm (7.0; 13.5) vs. 5.0 mm (3.0; 8.0), p < .001; lateral 8.0 mm (5.5; 11.0) vs. 5.0 mm (3.0; 7.0), p = .001) and showed a numerical difference in the RME (medial 11.5 mm (8.0; 21.5) vs. 10.0 mm (5.0; 21.0), p = .112; lateral 11.5 mm (8.0; 15.0) vs. 10.0 mm (5.0; 24.0), p = .744) and SSM-cohort (medial 13.0 mm (10.0; 20.0) vs. 8.0 mm (4.0; 12.0), p = .075; lateral 12.0 mm (10.0; 15.0) vs. 7.0 mm (4.0; 11.0), p = .068).

Discussion

RFGT after a mastectomy can deteriorate the prognosis of a breast cancer patient [7, 10, 16, 17, 21, 22]. Various studies – based on radiologic imaging [1,2,3,4] or pathology assessment [5,6,7,8,9,10,11,12,13,14,15,16,17,18] – have demonstrated high prevalence rates of RFGT. The present study aimed to identify risk factors for RFGT, that should be taken into consideration in the course of surgical treatment to potentially prevent or minimize the presence of RFGT.

In the present study higher RFGT volumes were observed following a NSM compared to a SSM and RME (p < .001). Notably, existing literature on risk factors of RFGT is contradictory (see Tables 5 and 6). While some studies also identified type of mastectomy as a risk factor for RFGT – with more RFGT being detected following a NSM than a SSM [3, 12, 27] than a total mastectomy [2] – other studies did not (MRI-based evaluation including NSM, SSM and simple mastectomy [20] and pathology-based assessment including total glandular mastectomy and modified radical mastectomy [14]).

Table 5 Evaluated risk factors for RFGT I
Full size table
Table 6 Evaluated risk factors for RFGT II
Full size table

Woitek R et al. [3] ascribed the higher amount of RFGT following a NSM compared to a SSM to the more linear incision in NSM and hence limited surgical accessibility in comparison to the circular incision around the areola in SSM.

Furthermore, the higher volume of RFGT following a NSM compared to other types of mastectomy can be explained by the preservation of the skin envelope and the NAC. Owing to the frequent irregularity or lack of the superficial fascia of the breast – which is considered as landmark of dissection [18, 28] – the identification of the correct anatomic dissection plane in the course of skin flap preparation is often impeded and results in incomplete FGT removal. Furthermore, the thickness of the subcutaneous fat tissue was shown to be inhomogeneous even within one breast [12] complicating tissue preparation and attributing to the higher likelihood of RFGT when the skin envelope is preserved.

Additionally, terminal duct lobular unit (TDLU) density was found to be higher in the NAC than in the adjacent skin [12] explaining the higher risk of RFGT if the NAC is preserved. High prevalence rates and RFGT location predominantly in the retroareolar area were reported by various authors [4, 6, 12, 20, 29, 30].

Notably, despite the higher likelihood of RFGT following NSM compared to any other mastectomy type, NSM in comparison to SSM or total mastectomy has shown no significant difference in recurrence rates or overall survival [31,32,33,34]. This highlights the significance of tumor biology rather than surgical techniques regarding the oncologic outcome of breast cancer patients.

In the present study RFGT volume was significantly associated with reconstruction (p = .012) – with higher RFGT volumes in breasts with immediate primary fixed volume implant reconstruction compared to immediate primary tissue expander reconstruction, delayed primary tissue expander reconstruction and no reconstruction, as well as ADM or mesh (p = .031) – with higher RFGT volumes in breasts reconstructed with an ADM compared to reconstruction with a synthetic mesh or no use of an ADM or synthetic mesh.

This might be explained by the aspiration to preserve a thicker skin envelope, which has been associated with RFGT in literature [1, 2, 4, 10, 17, 23, 24, 27], if implant-based reconstruction is performed in order to ensure flap perfusion and viability. Notably, Roy De Vita et al. [35] showed a statistically significant association between complications after NSMs and skin flaps of less than 5 mm. Frey JD et al. [36] identified a NSM flap thickness of less than 8.0 mm to be an independent predictor of ischemic complications.

Further procedure and surgeon-related risk factors for RFGT analysed in literature are displayed in Table 5.

Regarding patient related risk factors for RFGT, RFGT volume was significantly associated with patient age in the present study (p = .022) – with higher RFGT volumes found in younger patients. This might be explained by the high breast density in young patients (which has been associated with RFGT in literature [27]). Yet, contrary to the present study, Zippel D, et al. [1] detected more RFGT in older patients. Other authors found no association of RFGT with age at all [10,11,12].

RFGT volume was also associated with preoperative FGT volume in the present study (p = .012) – with higher RFGT volumes in case of higher FGT volumes – and preoperative whole breast volume (p = .030) – with higher RFGT volume in case of higher whole breast volume. Similar results were shown by Dietzel F, et al. [27]. This might be explained by the technically more difficult tissue preparation in larger breasts.

Further patient-related risk factors for RFGT studied in literature are displayed in Table 6.

Notably, contradictory findings of risk factors of RFGT in literature including the present study might be explained by methodological differences between the studies: MRI vs. pathology-based RFGT assessment and different RFGT sampling techniques in pathology -based studies (sample collection from the superficial dissection plane of the mastectomy specimen versus mastectomy cavity). In addition, the relatively small sample cohorts studied and the often retrospective study design also contributed to the heterogeneity of the results.

We furthermore, evaluated the tissue envelope thickness of the analysed breasts and found a reduction in the postoperative compared to preoperative skin and subcutaneous fat tissue thickness reaching statistical significance in the NSM cohort (medial p < .001, lateral p = .001) and showing a numerical difference in the RME and SSM-cohort. Notably, as skin flaps of less than 5 mm have been associated with ischemic complications [35], we suggest preoperative planning and determination of the patients’ individual target flap thickness – in accordance with Woitek R et al. [3] – in order to ensure thorough preservation of the skin and subcutaneous fat tissue thickness as well as complete removal of FGT.

Limitations of the present study include the retrospective study design and the partially missing data as well as the relatively small sample cohort.

Conclusions

In conclusion, identification of risk factors for RFGT is a first step to potentially prevent the remaining of FGT after mastectomy in the future. The described reduction in the post- compared to preoperative skin and subcutaneous fat tissue thickness should be avoided considering the known associated increase in risk for ischemic complications.