In our study including 257 women with stage ≤ 2A breast cancer who underwent CCM or EAPM and followed for up to 10 years, we found that EAPM performed significantly better than CCM in terms of cosmetic outcomes for location B at 2, 5, and 10 year-follow ups. Overall, the proportion of satisfactory cosmetic outcomes was significantly higher in EAPM than in CCM at 2-year follow-up but this was non-significant in subsequent follow-ups. Furthermore, in terms of the individual criteria of cosmetic outcomes, we found that EAPM was significantly superior to CCM in terms of atrophy and scar and this effect was consistently shown across all follow-ups (i.e., 2, 5, and 10 years), although atrophy did not reach statistical significance at 5-year follow-up (P = 0.058). Finally, survival analysis showed that there were no significant differences in terms of post-operative overall mortality, breast cancer-specific mortality, and recurrence between EAPM and CCM.
The finding that better cosmetic outcome overall in EAPM was only found at 2-year follow-up may be related to the filling of the dead space with an absorbable mesh. In EAPM, the dead space is filled with absorbable meshes plus reactive tissue fluid, two to three days after the injection of Vicryl mesh wrapped in Intershied. Foreign-body reaction occurs and promotes granulation and the formation of connective tissue along the margin of the dead space. One to two months post-operatively, the margin becomes suitably hard, and the dead space forms a comparatively fixed shape13)14). It is possible that the benefits of EAPM decreased after two years due to reduction in the fluid stored in the dead space, which may have led to asymmetry via traction of the surrounding tissue. It is difficult to clarify the reason for the fluid loss because pathological research cannot be performed. However, one of the causes may be insufficient margin formation (i.e., encapsulation), which plays an important role in the cosmetic outcome of EAPM.
Margin formation may also be affected by factors such as age, menopause status, BMI, tumor location, excision volume, the ease of granulation tissue formation, and mammary gland shape, as well as other external factors such as chemotherapy, endocrine therapy, and radiation therapy. These conditions are all different in each case, and thus, it is unknown which factor had the greatest impact on cosmetic outcome, especially considering that these factors may be intertwined and have synergistic effects. However, based on our results, in the case of EAPM, it seems to be clear that tumor location is a determinant of cosmetic outcome given that location B was significantly and consistently associated with approximately 3 times higher rates of satisfactory results in EAPM compared to CCM during the 10-year follow-up.
It is widely known that a lower tumor location is associated with postoperative cosmetic outcome difficulties in BCT15–17). However, EAPM performed much better than CCM in location B and this effect was long-lasting (i.e., observed for up to at least 10 years after the operation). This may be related to the fact that cases of location B usually have a smaller resected volume than in those of other locations in EAPM, and it may be easily fixed with the use of a brassiere after operation. Therefore, margin formation may have completed at an early stage with early stabilization of cosmetic outcomes. As for upper tumor location, although there were no significant differences in terms of cosmetic outcomes between EAPM and CCM across the years, when focusing only on EAPM, we found some difference between location A and C in terms of long-term outcomes. Specifically, of those who had overall satisfactory results at 2-year follow up, 33.3% (7/22) and 10.6% (5/47) had unsatisfactory results at 5-year follow up for location A and C, respectively, suggesting that location A is associated with worsening in cosmetic outcomes across time. A further examination into the individual criteria showed that a change from score 2 (satisfactory based on individual criteria) to score 0 or 1 (unsatisfactory) between 2 and 5-year follow-up in EAPM cases of location A and C occurred mainly for atrophy (A: 20.0%; n = 3/15, C: 20.0%; n = 7/35), ND (A: 11.1%; n = 2/18, C: 5.7%; n = 2/35), and BRA (A: 16.7%; n = 4/24, C: 2.0%; n = 1/49). As shown above, among patients who scored 2 for BRA at 2-year follow-up, a much larger proportion scored 0 or 1 at 5-year follow-up in location A as compared to location C. This may indicate that in location A, when the fluid in the dead space decreases, the accompanying traction of surrounding tissue is likely to affect the cosmetic outcome, but in location C, this effect may be minimal. However, given that this observation is derived from a small number of cases, this needs to be confirmed with larger studies. For the moment, we only consider this to be a possible characteristic associated with tumor location in EAPM. Finally, the fact that EAPM was superior to CCM in terms of scar and atrophy up to at least 10-years follow-up is likely to be explained by the periareolar incision where the scar is less noticeable, and the placing of buried subcutaneous sutures using 5 − 0 PDS-2 clear (see Appendix for details), while the insertion of the mesh may have led to less atrophy for certain locations (i.e., B and C).
There was no significant difference between CCM and EAPM in terms of all-cause mortality, breast cancer-specific mortality, and recurrence. Furthermore, we found no evidence of difference in the form of recurrence between the two procedures. Despite the fact that SSPM is associated with higher risk of local recurrence4,5), EAPM did not significantly increase local recurrence. We believe that this may be because (1) EAPM is not indicated for tumor size of > 3 cm, and cases with tumors fixed to the skin or muscle, and (2) a boost of 10 Gy or tumor bed 60Gy was added if pathological margins were close to the tumor edge (≤ 5mm). Fan et al18) reported that the indication of endoscopic subcutaneous mastectomy is limited to a distance of more than 0.5cm between the tumor surface and skin, determined by preoperative US, and our indication was similar. In addition, we judged that cases with cancer cells within 5 mm from the surface of the resected breast tissue were margin positive and added boost to all these cases. However, this is a criterion that is too strict in view of the current world trends19). We believe that this strict measure may have contributed to the reduction of local recurrence, but the possibility of overtreatment cannot be ruled out, and further studies are needed.
The results of our study showed that the beneficial effects of EAPM in terms of cosmetic outcomes is long-lasting but may be mainly restricted to location B. Whether there are any methods to improve cosmetic outcomes of EAPM for other locations is an area for future research. For example, given that one report suggested that the cosmetic outcome can be restored by injecting saline into the dead space when the fluid inside the dead space decreases18), we attempted to inject saline under guide of US at the 5th year in 5 EAPM cases of A and C areas which had experienced worsening of cosmetic outcome due to volume loss. However, in all cases, there was a complaint of strong pain before the injection volume exceeded 20 ml, and therefore, the procedure had to be interrupted. In addition, intraoperative US of these cases did not show clear stretching of the margin wall even during the injection of saline, and only showed that the entire dead space cavity swelled very slightly, while improvement in cosmetic outcome was not observed in any of these cases. Concrete conclusions cannot be drawn based on our experience of only 5 cases, but considering that cosmetic evaluation remained unchanged between 5 and 10 years, it is highly likely that the margin wall is basically fixed by this time. Our experience shows that at least after 5 years, it may be difficult to improve the cosmetic outcome by injecting additional saline, but there remains the possibility that this injection method may be beneficial at earlier timepoints.
It is also important to note that mesh infection is a complication of EAPM which may negatively affect cosmetic outcomes3). We implemented several countermeasures, but the rate of mesh infection did not change significantly. Currently, there are many operations that use absorbable meshes21–22), however, in EAPM, the mesh is placed directly in the dead space under the skin, and the reactive tissue fluid stored in the limited space may increase risk of infection.
Finally, a limitation of this study is that it was a single-center experience with relatively small sample size. Similar studies with larger sample size should be conducted to confirm our findings.