The effect of adjuvant radiotherapy on one- and two-stage prosthetic breast reconstruction and on autologous reconstruction: a multicenter Italian study among 18 Senonetwork Breast Centres

doi:10.21203/rs.3.rs-1712963/v1

Download PDF

Research Article

The effect of adjuvant radiotherapy on one- and two-stage prosthetic breast reconstruction and on autologous reconstruction: a multicenter Italian study among 18 Senonetwork Breast Centres

https://doi.org/10.21203/rs.3.rs-1712963/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Purpose: In modern breast cancer treatment we observed breast reconstruction growing role together with an extension in clinical indications for postmastectomy radiotherapy(PMRT). Optimal integration between reconstruction procedures and PMRT is still under debate. For this reason we performed a multicenter national study which aims to analyze the impact of PMRT on breast reconstruction.

Methods: we conducted a retrospective case-control multicenter study on women undergoing breast reconstruction. Data were collected from 18 Italian Breast Centres, and stored in a cumulative database which included: autologous reconstruction, direct-to-implant (DTI), and tissue expander/immediate (TE/I). For all patients, we described complications and surgical endpoints to complications such as reconstruction failure, explant, change in type of reconstruction and reintervention.

Results: From 2001 to April 2020, 3116 patients were evaluated. The risk for any complication was significantly increased in patients receiving PMRT (aOR, 1.73; 95% CI, 1.33‒2.24; p<0.001). PMRT was associated with a significant increase in the risk of capsular contracture in the DTI and TE/I groups (aOR, 2.24; 95% CI, 1.57‒3.20; p<0.001). Comparing type of procedures, the risk of failure (aOR, 1.82; 95% CI, 1.06‒3.12, p=0.030), explant (aOR, 5.49; 95% CI, 3.85‒7.83, p<0.001) and severe complications (aOR, 2.54; 95% CI, 1.88‒3.43, p<0.001) was significantly higher in the group undergoing DTI reconstruction as compared to TE/I reconstruction.

Conclusion: Our study confirms that autologous reconstruction is the one least impacted by PMRT while DTI appears to be the most impacted by PMRT if compared with TE/I which shows a lower rate of explant and reconstruction failure.

Trial registration number NCT04783818 and date of registration 1 March 2021 retrospectively registered

In the modern era of breast cancer treatment, the demand for postsurgery reconstruction is growing. For patients submitted to mastectomy, preserving the breast mound with immediate or delayed reconstruction is crucial to ensure an adequate quality of life. [1, 2]

Several reconstruction techniques are available: two-stage tissue expander and implants (TE/I), single-stage direct-to-implant reconstruction (DTI), and autologous tissue grafts.

Choosing the optimum type of reconstructive techinique represents a clinical challenge that depends on patient’s preference, balancing risks and benefits of each technique, baseline risk factors for reconstruction failure such as high BMI or smoking, and the need for postmastectomy radiation therapy (PMRT). [3]

Recently, in parallel with the increase of mastectomy and breast reconstruction rates, clinical indications for radiotherapy after mastectomy have been extended. [4, 5]

PMRT should be proposed to patients with locally advanced disease (T3 or higher) or with pathologically involved lymph nodes (N2-N3, N1 under discussion) [6, 7]. PMRT volumes includes: chest wall, regional lymph nodes, dissected / non-dissected armpit, supraclavicular fossa, and internal mammary nodes. [8]

The benefit of PMRT and the growing clinical practice of postmastectomy reconstruction have raised many questions about their optimal integration.

There are few studies, with significant numbers, evaluating the effect of radiotherapy on the three types of reconstruction. This represents significant gap in the published medical literature.

In particular, given the extreme variability in clinical approaches, there is no consensus regarding the most appropriate surgical technique or the correct reconstructive timing with respect to radiotherapy. [9]

Senonetwork Italia, a non-profit organization devoted to support the quality of multidisciplinary breast cancer (BC) care, has promoted the collection of data generated by several Italian breast centres of the network into a multicenter database. The present work aims to analyze the impact of PMRT on breast reconstruction and to compare the three principal types of reconstruction techniques evaluating the one associated with the lowest rate of complications in PMRT settings.

Sample and Data Collection

We conducted a retrospective case-control multicenter study on women undergoing breast reconstruction in 18 Italian centers from 2001 to 2020 (Listed in Fig. 1). The study was submitted for approval to the ethics committee of the Humanitas Research Hospital, and uploaded to Clinical Trials (ClinicalTrials.gov NCT04783818).

Data were collected from 18 Italian Breast Centres, and stored in a cumulative database created for this study and managed by the Humanitas Research Hospital.

Inclusion and exclusion criteria are listed in Fig. 2.

In the compilation of the database, each sample was considered as a breast and not as a patient; patients undergoing mastectomy and subsequent bilateral reconstruction were considered as two samples.

Reconstructive Procedure And Endpoints

In our study population we included: autologous reconstruction, direct-to-implant (DTI), and tissue expander/immediate (TE/I).

Patients who underwent reconstruction with latissimus dorsi and implant or expander were included either in DTI group or TE/I group.

In order to uniform the study population, in the TE/I group we decided to exclude patients who received radiotherapy on definitive implant since the majority of patients received radiotherapy during the tissue expander procedure.

In the prosthetic group we only included retropectoral reconstruction since the prepectoral patients did not demonstrate an adequate follow-up. In addition only a few patients had an acellular dermal matrix reconstruction (ADM) and we decided to exclude them .

For all patients, we described complications and possible surgical endpoints to complications.

We considered as mild complications: hematoma, seroma, skin necrosis, addominal and lumbar pain, liponecrosis and volume loss. On the other hand severe complications included: capsular contraction, hernia, flap necrosis, bulging, microvascular complication and late infection.

We divided infections in early and late, considering late infections as occurring 30 or more days from surgery.

We did not grade capsular contracture in order to overcome heterogeneity of evaluation; we considered capsular contracture as present or absent.

Four different endpoints for reconstructive complications were chosen: reconstruction failure, explant, change in type of reconstruction and reintervention.

The term “reconstruction failure” is referred to patients that do not have any other breast reconstruction after implant or flap removal. The term “explant”, on the other hand, was employed when the prosthesis or flap were removed without contraindication for another reconstructive procedure.

The change in type of reconstruction was reported as when after the explant, the primary reconstruction (DTI, TE/I or flap) was replaced by, or integrated with, another type of reconstruction.

Reinterventions included: escarectomy in cases of necrosis, seroma and hematoma drainage, capsular revision or implant substitution, and microvascular revision.

Statistical analysis

We described patients’ characteristics and treatments received at baseline, as well as postmastectomy outcomes and complications, in the overall cohort and by type of surgical procedure performed.

We summarized data as medians and interquartile ranges (IQR) or percentages, and compared them across groups of patients by using the chi-squared test or Fischer exact test (categorical data) or the Kruskal-Wallis test (continuous variables) as appropriate.

We investigated the risk of failure, explant and reintervention, as well as risk of any complication, severe complications and any infection, according to the type of procedure performed by means of multivariable hierarchical logistic regression.

To investigate the association between PMRT and key negative outcomes and complications, we performed multivariable hierarchical logistic regression analyses and specified a random intercept for each of the 18 recruiting centers. The model was pre-specified and included relevant confounders identified by literature screening and expert opinion. The analyses were adjusted for age, BMI, smoking habits, diabetes, other autoimmune diseases, type of chemotherapy, axillary dissection, and type of surgical procedure. Missing were less than 2% per variable, with the exception of BMI (13.5%). We considered values as missing at random, and performed a complete case analysis. However, for each outcome measure, we performed a regression analysis not including BMI (Table 3 model 1), and a maximally adjusted analysis including BMI (Table 3 model 2).

We used STATA 17.0 (Stata Corp., College Station, TX, USA) for the statistical analysis. A two-sided p-value < 0.05 was considered statistically significant.

From 2001 to April 8, 2021, 3116 records were retained for the analysis: 187 autologous, 1227 direct-to-implant, and 1702 tissue expander/immediate reconstructions. (Fig. 2)

Baseline demographic and clinical characteristics of the overall cohort stratified by surgical procedure are reported in Table 1.

Table 1

Baseline demographic, clinical characteristics and treatments of the overall cohort and stratified by surgical procedure and post-mastectomy radiotherapy.
Variable	Autologous reconstruction (187 pts)		Direct to implant (1,227 pts)		Tissue expander/immediate (1,702 pts)		Overall cohort†
Variable	PMRT	NO PMRT	PMRT	NO PMRT	PMRT	NO PMRT	Overall cohort†
Number of patients (n = 3,116)	79 (42.3)	108 (57.7)	392 (31.9)	835 (67.9)	696 (40.9)	1,006 (59.1)	3,116
Age (years; n = 3,102)	50 (45‒59)	50.5 (45‒56)	49 (43‒56)	47 (41‒53)	49 (43‒57)	50 (44‒58)	49 (43‒56)
BMI (kg/m²; n = 2,698)	26.5 (24.0‒28.4)	25.0 (23.0‒27.1)	22.4 (20.6‒24.2)	22.0 (20.1‒23.7)	23.7 (21.0‒26.6)	23.2 (21.1‒25.8)	22.9 (20.8‒25.3)
≥ 30	11 (15.3)	9 (9.2)	17 (4.6)	11 (1.4)	54 (9.2)	61 (7.6)	163 (6.0)
Follow-up (years; n = 2,832)	2.6 (0.9‒3.4)	2.2 (0.5‒2.8)	2.8 (2.0‒4.0)	2.4 (1.9‒3.3)	2.3 (1.5‒3.4)	2.4 (1.6‒3.2)	2.4 (1.6‒3.3)
Smoking (n = 3,077)
Smokers	12 (16.4)	6 (5.7)	57 (15.2)	127 (15.4)	114 (16.5)	131 (13.0)	447 (14.5)
Ex-smokers	5 (6.9)	13 (12.4)	53 (14.1)	91 (11.1)	105 (15.2)	105 (10.5)	372 (12.1)
Non-smokers	56 (76.7)	86 (81.9)	265 (70.7)	606 (73.5)	473 (68.3)	769 (76.5)	2,258 (73.4)
Diabetes mellitus (n = 3,086)	5 (6.3)	1 (0.9)	9 (2.3)	25 (3.1)	31 (4.5)	31 (3.1)	102 (3.3)
Other autoimmune disorders (n = 3,119)	8 (10.1)	6 (5.6)	31 (7.9)	81 (9.7)	43 (6.2)	60 (6.0)	229 (7.3)
Chemotherapy (n = 3,115)
No	8 (10.1)	37 (34.3)	41 (10.5)	444 (53.4)	139 (20.0)	554 (54.1)	1,213 (38.9)
Adjuvant	46 (58.2)	53 (49.1)	225 (57.4)	219 (26.4)	372 (53.5)	355 (35.3)	1,273 (40.9)
Neoadjuvant	25 (31.7)	18 (16.6)	126 (32.1)	168 (20.2)	175 (25.1)	104 (10.3)	616 (19.8)
Adjuvant and neoadjuvant	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	10 (1.4)	3 (0.3)	13 (0.4)
Axillary dissection (n = 3,112)	61 (77.2)	33 (30.6)	322 (82.1)	182 (21.9)	545 (78.3)	292 (29.1)	1,435 (46.1)
Values are expressed as absolute frequency (percentage) for categorical variables and as median (interquartile range) for continuous variables.
Abbreviation
BMI, body mass index; PMRT, post-mastectomy radiotherapy.

The median age of patients was 49 years (IQR, 43‒56), with patients undergoing DTI being a median of 3 years younger than those receiving other types of surgical procedures, Table 1. The median BMI was 22.9 (IQR, 20.8‒25.3); it was significantly higher in patients undergoing autologous reconstruction (p < 0.001). This was reflected by the higher proportion of obese patients in this group (p < 0.001, Supplementary Table 1). Of our patients, 14.5% of were current smokers, 3.3% had diabetes and 7.3% had another autoimmune disease. Most patients received chemotherapy (61.1%), 46.1% underwent axillary dissection, and 37.5% PMRT. The proportion of patients undergoing these treatments and procedures significantly differed by type of surgical procedure performed, Table 1.

Postsurgical Outcomes And Complications

The key negative postoperative outcomes and complications by type of surgical procedure are reported in Table 2 and Supplementary Table 2. Failure occurred in a small proportion of patients (2.3%), with no significant differences between patients undergoing different types of reconstruction (p = 0.54).

Table 2

Post-operative outcomes and complications stratified by surgical procedure and post-mastectomy radiotherapy.
Variable	Autologous reconstruction (187 pts)		Direct to implant (1,227 pts)		Tissue expander/immediate (1,702 pts)		Overall cohort (3,116 pts)
	PMRT	NO PMRT	PMRT	NO PMRT	PMRT	NO PMRT	Overall cohort (3,116 pts)
Outcomes
Failure (no reconstruction; n = 3,113)	1 (1.3)	3 (2.8)	22 (5.6)	11 (1.3)	26 (3.8)	9 (0.9)	72 (2.3)
Change in type of reconstruction (n = 3,038)	9 (11.5)	0 (0.0)	10 (2.7)	4 (0.5)	7 (1.1)	0 (0.0)	30 (1.0)
Explant (n = 3,113)	4 (5.1)	3 (2.8)	106 (27.0)	67 (8.0)	88 (12.8)	48 (4.8)	316 (10.2)
Reintervention (n = 3,119)	3 (3.8)	3 (2.8)	17 (4.3)	11 (1.3)	253 (36.4)	70 (7.0)	357 (11.5)
Complications
Any complication (n = 3,118)	37 (46.8)	36 (33.3)	253 (64.5)	215 (25.8)	409 (58.9)	267 (26.5)	1,218 (39.1)
Severity of complications (n = 3,119)¹
No complication	42 (53.1)	72 (66.7)	139 (35.5)	620 (74.3)	286 (41.1)	739 (73.5)	1,900 (60.9)
Exclusively mild complications	16 (20.3)	17 (15.7)	22 (5.6)	69 (8.3)	62 (8.9)	135 (13.4)	321 (10.3)
Exclusively severe complications	3 (3.8)	6 (5.6)	170 (43.4)	83 (9.9)	256 (36.8)	80 (8.0)	598 (19.2)
Mild and severe complications	16 (20.3)	10 (9.3)	52 (13.3)	42 (5.0)	73 (10.5)	24 (2.4)	217 (7.0)
Severity unknown	2 (2.5)	3 (2.8)	9 (2.3)	21 (2.5)	19 (2.7)	28 (2.8)	83 (2.6)
Hematoma (n = 3,115)	7 (8.9)	7 (6.5)	13 (3.3)	21 (2.5)	24 (3.5)	39 (3.9)	111 (3.6)
Seroma (n = 3,114)	2 (2.5)	4 (3.7)	47 (12.0)	52 (6.2)	86 (12.4)	87 (8.7)	278 (8.9)
Cutaneous necrosis (n = 3,114)	18 (22.8)	10 (9.3)	26 (6.6)	51 (6.1)	35 (5.0)	53 (5.3)	193 (6.2)
Any infection (n = 3,115)	9 (11.4)	3 (2.8)	18 (4.6)	25 (3.0)	48 (6.9)	41 (4.1)	144 (4.6)
Timing of infections (n = 3,115)
No infection	70 (88.6)	105 (97.2)	374 (95.4)	810 (97.0)	644 (92.5)	965 (95.9)	2,971 (95.4)
Early infection	4 (5.1)	2 (1.9)	3 (0.8)	5 (0.6)	17 (2.4)	11 (1.1)	42 (1.3)
Late infection	2 (2.5)	1 (0.9)	12 (3.1)	13 (1.6)	15 (2.2)	15 (1.5)	58 (1.9)
Early and late infection	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	1 (0.1)	0 (0.0)	1 (0.0)
Timing unknown	3 (3.8)	0 (0.0)	3 (0.8)	7 (0.8)	19 (2.7)	15 (1.5)	47 (1.5)
Capsular contraction (n = 2,875)	NA	NA	203 (51.8)	104 (12.5)	295 (46.2)	72 (7.2)	674 (23.4)
Implant exposure (n = 2,928)	NA	NA	15 (3.8)	16 (1.9)	26 (3.8)	12 (1.2)	69 (2.4)
Implant rupture (n = 2,928)	NA	NA	2 (0.5)	3 (0.4)	12 (1.7)	13 (1.3)	30 (1.0)
Liponecrosis (n = 177)	9 (12.3)	10 (9.6)	NA	NA	NA	NA	19 (10.7)
Volume loss (n = 177)	11 (15.1)	4 (3.9)	NA	NA	NA	NA	15 (8.5)
Pain (n = 177)	2 (2.7)	2 (1.9)	NA	NA	NA	NA	4 (2.3)
Abdominal pain	2 (2.7)	2 (1.9)	NA	NA	NA	NA	4 (2.3)
Lumbar pain	0 (0.0)	0 (0.0)	NA	NA	NA	NA	0 (0.0)
Hernia (n = 177)	1 (1.4)	2 (1.9)	NA	NA	NA	NA	3 (1.7)
Flap necrosis (n = 177)			NA	NA	NA	NA
No	60 (82.2)	96 (92.3)	NA	NA	NA	NA	156 (88.1)
Partial	10 (13.7)	6 (5.8)	NA	NA	NA	NA	16 (9.1)
Full-thickness	3 (4.1)	2 (1.9)	NA	NA	NA	NA	5 (2.8)
Bulging (n = 177)	1 (1.4)	6 (5.8)	NA	NA	NA	NA	7 (4.0)
Microvascular complications (n = 177)²	6 (8.2)	4 (4.8)	NA	NA	NA	NA	11 (6.2)
Values are expressed as absolute frequency (percentage) for categorical variables and as median (interquartile range) for continuous variables.
Footnotes
¹ Mild complications: hematoma, seroma, cutaneous necrosis, early infection, abdominal or lumbar pain, liponecrosis, volume loss. Severe complications: late infection, capsular contraction, implant exposure, implant rupture, hernia, flap necrosis, bulging, microvascular complications requiring surgery.
² Microvascular complication requiring surgery.

It should be noted that in irradiated expander, failure accounted for 3.8% of the patients while in irradiated DTI this increased to 5.6%.

Accordingly, the proportion of patients undergoing explant was significantly higher in the DTI group (14.1%; p < 0.001, Supplementary Table 2); taking irradiation into account, we observed an increase in explant rate comparing expander and DTI (both irradiated and non-irradiated) which resulted as: 12.8% vs. 27%, and 4.8 vs. 8% respectively, Table 2.

Table 3

Multivariable hierarchical logistic regression analyses investigating the association between post-mastectomy radiotherapy and key post-surgical outcomes and complications. Each estimate is adjusted for age, smoking habits, diabetes mellitus, other autoimmune diseases, type of chemotherapy, axillary dissection and type of surgical procedure performed (autologous reconstruction, direct to implant or tissue expander/immediate). Estimates in the second column (multivariable model 2) are additionally adjusted for BMI.
Dependent variable	Radiotherapy
	Multivariable model 1		Multivariable model 2
	aOR (95% CI)	P-value^†	aOR (95% CI)	P-value^†
Outcomes
Failure	2.60 (1.38‒4.88)	0.003	2.72 (1.39‒5.32)	0.004
Change in type of reconstruction	5.66 (1.20‒26.6)	0.028	7.99 (1.36‒47.0)	0.021
Explant	2.31 (1.40‒3.80)	0.001	2.49 (1.49‒4.15)	< 0.001
Reintervention	1.24 (0.79‒1.94)	0.347	1.49 (0.89‒2.48)	0.128
Complications (all procedures)
Any complication	2.00 (1.58‒2.54)	< 0.001	1.73 (1.33‒2.24)	< 0.001
Severe complication¹	2.80 (2.09‒3.74)	< 0.001	2.17 (1.58‒2.97)	< 0.001
Mild complication²	1.60 (1.18‒2.17)	0.003	1.28 (0.91‒1.80)	0.162
Hematoma	0.61 (0.33‒1.14)	0.120	0.48 (0.23‒1.02)	0.056
Seroma	1.82 (1.23‒2.70)	0.003	1.49 (0.92‒2.41)	0.101
Cutaneous necrosis	0.91 (0.60‒1.39)	0.664	0.68 (0.42‒1.10)	0.114
Any infection	1.76 (1.09‒2.86)	0.022	1.91 (1.06‒3.42)	0.030
Late infection	1.18 (0.52‒2.64)	0.693	0.82 (0.28‒2.38)	0.713
Complications in DTI and TE/I groups
Capsular contraction	3.18 (2.31‒4.37)	< 0.001	2.24 (1.57‒3.20)	< 0.001
Implant exposure	2.05 (1.07‒3.92)	0.029	1.94 (0.99‒3.79)	0.054
Implant rupture	1.31 (0.49‒3.54)	0.591	1.85 (0.64‒5.36)	0.258
Values are expressed as absolute frequency (percentage) for categorical variables and as median (interquartile range) for continuous variables. Due to the small sample size and number of events in the group undergoing autologous reconstruction, it was not possible to perform the multivariable regressions using as outcomes the following complications occurring exclusively in this group: liponecrosis, volume loss, pain, hernia, flap necrosis, bulging, microvascular complications.
Footnotes
† Chi-squared test for categorical variables; Kruskal-Wallis test for continuous variables.
¹ Severe complications: late infection, capsular contraction, implant exposure, implant rupture, hernia, flap necrosis, bulging, microvascular complications requiring surgery.
² Mild complications: hematoma, seroma, cutaneous necrosis, early infection, abdominal or lumbar pain, liponecrosis, volume loss.

It should be noted that the proportion of patients needing a reintervention was higher in the TE/I group (19.0%; p < 0.001), when compared with both autologous reconstruction and DTI.

The number of patients who needed a change in the type of reconstruction was higher in the group undergoing autologous reconstruction (4.9%; p < 0.001, Supplementary Table 2) and more precisely in the irradiated autologous reconstructions (11.5% against 0% of the non-irradiated autologous group, Table 2).

In 39.1% of the cases we registered at least one complication divided as follows: 19.2% severe complication, 10.3% mild complication, 7.0% concomitant mild and severe complications, and 2.6% of unknown or unclassified severity. Infections occurred in 4.6% of cases (n = 144), Supplementary Table 2.

For implant-based breast reconstruction, the most frequent complication was capsular contracture accounting for 12.5% of patients undergoing DTI and 7.2% in TE/I in non-irradiated patients. Such percentages increased dramatically in irradiated patients up to 51.8% and 46.2% respectively, Table 2.

In the group undergoing autologous reconstruction the most common severe complications were: presence of flap necrosis of any grade (11.9%), microvascular complications requiring reintervention (6.2%), and bulging (4.0%). While in the groups undergoing DTI and TE/I reconstruction the most severe complication was by far capsular contracture occurring respectively in 25.0% (n = 307) and 22.3% (n = 367) of the reconstructions performed, Supplementary Table 2.

Postmastectomy Radiotherapy As A Risk Factor For Key Outcomes And Complications

The multivariable-adjusted odds ratios (i.e., adjusted for age, BMI, smoking habits, diabetes, other autoimmune diseases, type of chemotherapy, axillary dissection and type of surgical procedure performed) between PMRT and key outcomes and complications are reported in Table 3.

When taking into account the above-mentioned confounders, PMRT increased the risk of key negative outcomes including reconstruction failure (aOR, 2.72; 95% CI, 1.39‒5.32; p = 0.004), need for a change in type of reconstruction (aOR, 7.99; 95% CI, 1.36‒47.0; p = 0.021), explant (aOR, 2.49; 95% CI, 1.49‒4.15; p < 0.001), but not reintervention (aOR, 1.49; 95% CI, 0.89‒2.48; p = 0.128).

The risk for any complication was significantly increased in patients receiving PMRT (aOR, 1.73; 95% CI, 1.33‒2.24; p < 0.001), in particular those of severe grade (aOR, 2.17; 95% CI, 1.58‒2.97; p < 0.001), while mild complications were not significantly affected by irradiation. PMRT was associated with a significant increase in the risk of any infection (aOR, 1.91; 95% CI, 1.06‒3.42; p = 0.030), but not of late infections (aOR, 0.82; 95% CI, 0.28–2.38; p = 0.713) together with an association with the risk of capsular contracture in the DTI and TE/I groups (aOR, 2.24; 95% CI, 1.57‒3.20; p < 0.001).

Risk Of Negative Outcomes And Complications By Type Of Procedure

In adjusted analyses taking into account: PMRT, age, BMI, smoking habits, diabetes, other autoimmune diseases, type of chemotherapy and axillary dissection, the risk of failure (aOR, 1.82; 95% CI, 1.06‒3.12, p = 0.030), explant (aOR, 5.49; 95% CI, 3.85‒7.83, p < 0.001), any complication (aOR, 1.83; 95% CI, 1.43‒2.34, p < 0.001), and severe complications (aOR, 2.54; 95% CI, 1.88‒3.43, p < 0.001) was significantly higher in the group undergoing DTI reconstruction as compared to TE/I reconstruction. Although the adjusted odds of reintervention were about 20 times lower in patients with DTI reconstruction (aOR, 0.05; 95% CI, 0.03‒0.08, p < 0.001), Table 4.

Table 4

Multivariable hierarchical logistic regression analyses investigating the association between type of surgical procedure performed and key negative outcomes and complications. Each estimate is adjusted for post-mastectomy radiotherapy, age, BMI, smoking habits, diabetes, other autoimmune diseases, type of chemotherapy, and axillary dissection.
Type of surgical procedure	Failure		Explant		Reintervention		Any complication		Severe complications		Any infection
Type of surgical procedure	aOR (95% CI)	P	aOR (95% CI)	P	aOR (95% CI)	P	aOR (95% CI)	P	aOR (95% CI)	P	aOR (95% CI)	P	aOR (95% CI)	P
First model specification¹
Autologous reconstruction	ref.	‒	ref.	‒	ref.	‒	ref.	‒	ref.	‒	ref.	‒	ref.	‒
DTI reconstruction	1.75 (0.58‒5.27)	0.323	3.80 (1.49‒9.63)	0.005	0.11 (0.03‒0.43)	0.002	0.81 (0.51‒1.28)	0.360	1.11 (0.61‒2.02)	0.740	0.60 (0.25‒1.41)	0.242	0.12 (0.03‒0.49)	0.004
TE/I reconstruction	0.96 (0.32‒2.86)	0.940	1.14 (0.44‒2.91)	0.792	2.20 (0.59‒8.17)	0.240	0.44 (0.27‒0.71)	0.001	0.44 (0.24‒0.80)	0.008	0.77 (0.32‒1.83)	0.552	0.05 (0.10‒0.27)	< 0.001
Second model specification¹
DTI reconstruction	ref.	‒	ref.	‒	ref.	‒	ref.	‒	ref.	‒	ref.	‒	ref.	‒
Autologous reconstruction	0.57 (0.19‒1.73)	0.323	0.26 (0.10‒0.67)	0.005	9.09 (2.30‒35.9)	0.002	1.24 (0.78‒1.98)	0.360	0.90 (0.50‒1.64)	0.740	1.67 (0.71‒3.93)	0.242	8.59 (2.02‒36.5)	0.004
TE/I reconstruction	0.55 (0.32‒0.94)	0.030	0.30 (0.22‒0.41)	< 0.001	20.0 (12.1‒33.0)	< 0.001	0.55 (0.43‒0.70)	< 0.001	0.39 (0.29‒0.53)	< 0.001	1.28 (0.74‒2.23)	0.378	0.45 (0.16‒1.26)	0.129
Third model specification¹
TE/I reconstruction	ref.	‒	ref.	‒	ref.	‒	ref.	‒	ref.	‒	ref.	‒	ref.	‒
Autologous reconstruction	1.04 (0.35‒3.11)	0.940	0.88 (0.34‒2.25)	0.792	0.45 (0.12‒1.69)	0.240	2.27 (1.41‒3.67)	0.001	2.29 (1.24‒4.23)	0.008	1.30 (0.55‒3.08)	0.552	19.1 (3.75‒97.6)	< 0.001
DTI reconstruction	1.82 (1.06‒3.12)	0.030	3.34 (2.46‒4.54)	< 0.001	0.05 (0.03‒0.08)	< 0.001	1.83 (1.43‒2.34)	< 0.001	2.54 (1.88‒3.43)	< 0.001	0.78 (0.45‒1.36)	0.378	2.23 (0.79‒6.27)	0.129
Footnotess:
¹ We have specified different baseline levels for the variable indicating the type of surgical procedure performed and have alternatively selected a different group as the one corresponding to the baseline risk. This allowed to show all the formal comparisons across groups.

The group of patients undergoing DTI reconstruction had a significantly higher risk of explant (aOR, 3.03; 95% CI, 1.15‒8.03, p = 0.026) as compared with those who underwent autologous reconstruction, but the adjusted odds of reinterventions were about nine times lower (aOR, 0.11; 95% CI, 0.03‒0.43, p = 0.002). As compared to those with a TE/I reconstruction, those undergoing autologous reconstructions had an increased risk of complications (aOR, 2.27; 95% CI, 1.41‒3.67, p = 0.001) and severe complications (aOR, 2.29; 95% CI, 1.24‒4.23, p = 0.008), Table 4.

Patients undergoing autologous reconstruction had a higher risk of changing the type of reconstruction as compared to those undergoing DTI and TE/I respectively aOR 0.12;95% CI, 0.03–0.49 and aOR 0.05; 95% CI, 0.10–0.27.

There were no significant differences in terms of infections across any of the three groups.

Nowadays indications to PMRT are increasing [4, 5] together with the rate of breast reconstruction given an increased awareness of its fundamental role.

In order to guide physicians to a better practice, there is a clear need for guidelines and position statements in the medical-scientific community. In particular, the impact of PMRT on breast reconstruction has been a subject of longstanding research, but a large study population is still needed to confirm the evidence. [10, 11]

For this reason, Senonetwork Italia supported this project in order to evaluate the PMRT effect on breast reconstruction and to compare the three principal types of reconstruction techniques.

The retrospective experience of 18 centers was collected and the different clinical experiences reported were collected and examined to account for similarities and differences. In the three different populations we analyzed the percentage of absolute reconstruction failure, explant, reintervention and change in type of reconstruction.

Categorizing reintervention represented a particularly challenging task, since we included in this group general surgical complications such as escarectomy in cases of necrosis, seroma or hematoma drainage together with implant-related reinterventions and microvascular revisions. Analyzing reintervention was aimed at obtaining data regarding the possibility to receive additional intervention after primary surgery, thus offering patients comprehensive preoperative information.

Our study confirms the negative effect of PMRT on breast reconstruction after adjusting for a consistent set of clinical and surgical confounders. PMRT increased the risk of complication, explant and reconstruction failure - but interestingly - not of late infection. In addition to that, our analysis confirmed that radiotherapy is related to a dramatic increase in the risk for capsular contraction in both DTI and TE/I patients .

In our study population, we collected a great majority of implant-based breast reconstruction (94%) including patients submitted to both latissimus dorsi and implant, such data are consistent with Italian experience where prosthetic reconstruction is prevalent.

Nevertheless, autologous reconstruction has been adopted as a viable option in view of adjuvant radiotherapy, based on the widespread belief of the better safety of this technique, especially in the case of PMRT or as a salvage procedure after implant removal. [12, 13].

In our population, absolute failure rate was as low as 2.3%, with no significant differences between type of reconstruction. However, when taking into consideration radiotherapy, autologous reconstruction appeared to be the less impacted if compared with implant based breast reconstruction.

Our observations, although limited by the reduced percentage of patients, are consistent with more recent studies that confirm the superiority of autologous reconstruction in terms of complication rates and reconstructive failure in case of PMRT. [14–19]

Despite the advantages described above, the autologous approach is characterized by longer intervention and recovery times than prosthetic reconstructions together with morbidity of the donor site. [18] In our analysis, an additional disadvantage characterizing the autologous approach was the increase in the risk of undergoing a change in the type of reconstruction with respect to prosthetic reconstruction, Table 4.

In our cohort, we observed that the risk of a change in type of reconstruction was greatly increased in patients undergoing autologous reconstruction versus DTI and TE/I reconstructions. To our knowledge, this is the first study to propose such an analysis exploring how the type of reconstruction is affected by complications.

It should be noted that this event happened rarely in the overall cohort (30 cases,1% of the total), and for autologous cases flap conversion was carried out in 8 out of 9 cases with autologous fat grafting, and in one case with latissimus dorsi and prosthesis so that the main choice was to maintain the autologous nature of reconstruction.

Nevertheless, such an observation was highly statistically significant due to the large relative difference across groups and the large overall sample of this multicentric cohort and confirms that an accurate selection of candidate patients is mandatory together with the selection of highly specialized centres [20].

Given the safer profile of autologous reconstruction, the greatest dilemma in prosthetic breast reconstruction is whether to prefer DTI or TE/I reconstruction, especially when radiotherapy is administered. There is a growing debate about which is the preferable implant-based breast reconstruction to offer in case of PMRT. [8, 14–15, 17]

Lin and colleagues argued in favor of DTI, comparing irradiated single stage breast reconstruction and two-stage breast reconstruction when TE is irradiated after the first stage [21].

The multivariate analysis of Naoum et al. showed a significant association between TE/I and higher risk of infection and implant failure compared to AR, while single-stage-direct-to-implant and AR were comparable, also when neoadjuvant chemotherapy was performed. [14, 22]

Unlike Lin and Naoum’s reports, in our study we observed a higher proportion of explant in irradiated DTI when compared with irradiated TE .

In particular, a more detailed comparison between our analysis and that of Naoum and coworkers reveals that the two analyses differ slightly. In our analysis, radiotherapy is considered as a confounder, while in Naoum’s paper non-irradiated patients are excluded. Indeed, in our analysis, results are applicable to patients both with and without radiotherapy, as radiotherapy was included in the multivariable model.

We are convinced that our statistical analysis is more consistent with current clinical practice, in which the need for PMRT is most of the time not known before breast reconstruction, so we aim to evaluate the best reconstruction procedure considering any clinical variable. Indeed, the statistical adjustment performed on our population, which allowed to include both patients who underwent PMRT and those who did not, demonstrated a higher power in detecting any difference when compared to the analysis which takes into account irradiated patients only.

According to our data, TE/I reconstruction appears to be safer when compared to DTI; indeed in patients who underwent direct implant we observed that complications may lead more frequently to failure and explant, while the positioning of implants after expander substitution seemed to lead to failure rate being halved both in irradiated and non-irradiated fields.

Mun’s metanalysis [23] presented comparable results based on the belief that direct pressure or excess tension on a mastectomy skin flap due to insertion of a large fixed-volume implant could increase the rate of reconstruction failure and explant, especially in the case of PMRT.

Flap necrosis and associated complications such as seroma and infection may lead in TE/I to a higher rate of reintervention, as we observed in our case series. Furthermore, we are convinced that prosthetic complications impact DTI and TE/I equally, especially radiation-related complications. However, such an impact, in TE/I breast reconstruction is not reflected in an increase in failure or explant but can result in an increased need for implant substitution or capsular revision that could also justify the increase in reintervention rate.

Although TE/I reconstruction appears to be the safest option when PMRT is administered, these observations warrant further evaluation to analyze the best timing of radiotherapy in relation to expander substitution, percentage of expander filling during radiotherapy, or whether the use of lipofilling during the substitution exerts an impact on final reconstruction failure.

Our analysis further highlights how radiotherapy leads to capsular contracture in both DTI and TE/I reconstruction. Such an observation based on a well-populated national database testifies the need for procedures to reduce and treat capsular contracture.

Our study is not without several limitations. First of all, it is retrospective in nature. Secondly, although collecting experience from 18 breast centers allowed us to obtain a substantial study population, there is some degree of variability in experience and clinical practice across centers. To overcome this limitation, we employed a multivariable hierarchical logistic regression analysis with a random intercept which took into account the fact that observations belonging to each recruiting center are to a certain degree correlated.

As previously stated, in order to obtain an additional evaluation, we analyzed the proportion of

reinterventions including a wide range of different procedures. We are aware that merging different procedures may somewhat mislead in terms of conclusions; nevertheless considering each single reintervention separately would not have allowed us to draw a meaningful comparison between the three main types of procedures because of a lack of statistical power.

We are convinced, however that this data still provides useful information regarding the chance to receive further procedures after the first operation.

Lastly, we could not include data regarding ADM usage. Such data is widely adopted in other studies, especially those conducted in the US – however, Italian centres do not routinely use ADM. Regarding prepectoral reconstruction, which has widespread indications, results published to date are limited by a short follow-up duration. We plan to conduct a more extensive, comprehensive and definitive analysis in the future.

Our large multicenter study confirms that autologous reconstruction is the one least impacted by PMRT. Nevertheless, it is more invasive, it needs a dedicated microsurgical team and it shows a higher risk of changing the type of reconstruction.

Direct-to-implant reconstruction appears to be the most impacted by PMRT if compared with the two-stage approach which shows a lower rate of implant explant and reconstruction failure but an increased risk for reintervention.

Choosing the most suitable reconstructive type in mastectomy patients who need PMRT remains a challenge for the surgeon.

Although the ultimate failure rate is low with all types, there seems to be no absolute best technique. It is essential that before each intervention the pros and cons of each technique are weighed, and that they are discussed with the patient.

We declare that:

All authors have participated in (a) conception and design, or analysis and interpretation of the data; (b) drafting the article or revising it critically for important intellectual content; and (c) approval of the final version.
This manuscript has not been submitted to, nor is under review at, another journal or other publishing venue.
The authors have no affiliation with any organization with a direct or indirect financial interest in the subject matter discussed in the manuscript
The authors declare that they have no conflict of interest.

Al-Ghazal SK, Fallowfield L, Blamey RW (2000) Comparison of psychological aspects and patient satisfaction following breast conserving surgery, simple mastectomy and breast reconstruction. Eur J Cancer. Oct;36(15):1938-43. DOI: 10.1016/s0959-8049(00)00197-0. PMID: 11000574
Cordeiro PG (2008) Breast reconstruction after surgery for breast cancer. N Engl J Med. Oct 9;359(15):1590 – 601. DOI: 10.1056/NEJMct0802899. PMID: 18843123
Ho AY, Hu ZI, Mehrara BJ, Wilkins EG (2017) Radiotherapy in the setting of breast reconstruction: types, techniques, and timing. Lancet Oncol. Dec;18(12):e742-e753. DOI: 10.1016/S1470-2045(17)30617-4. PMID: 29208440
Early Breast Cancer Trialists' Collaborative Group (EBCTCG) (2005) Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomised trials. Lancet. May 14–20;365(9472):1687 – 717. DOI: 10.1016/S0140-6736(05)66544-0. PMID: 15894097
Overgaard M, Hansen PS, Overgaard J, Rose C, Andersson M, Bach F, Kjaer M, Gadeberg CC, Mouridsen HT, Jensen MB, Zedeler K(1997) Postoperative radiotherapy in high-risk premenopausal women with breast cancer who receive adjuvant chemotherapy. Danish Breast Cancer Cooperative Group 82b Trial. N Engl J Med. Oct 2;337(14):949 – 55. DOI: 10.1056/NEJM199710023371401. PMID: 9395428
Recht A, Comen EA, Fine RE, Fleming GF, Hardenbergh PH, Ho AY, Hudis CA, Hwang ES, Kirshner JJ, Morrow M, Salerno KE, Sledge GW Jr, Solin LJ, Spears PA, Whelan TJ, Somerfield MR, Edge SB(2016) Postmastectomy Radiotherapy: An American Society of Clinical Oncology, American Society for Radiation Oncology, and Society of Surgical Oncology Focused Guideline Update. J Clin Oncol. Dec 20;34(36):4431–4442. DOI: 10.1200/JCO.2016.69.1188. PMID: 27646947
Heller DR, Zhuo H, Zhang Y, Parikh N, Fusi S, Alperovich M, Lannin DR, Higgins SA, Avraham T, Killelea BK(2021) Surgical Outcomes of Mastectomy with Immediate Autologous Reconstruction Followed by Radiation. Ann Surg Oncol. Apr;28(4):2169–2179. DOI: 10.1245/s10434-020-09122-0. PMID: 32974699
Agafonoff S, Kundu N, Schwarz G, Shah C(2020) Immediate Implant Reconstruction in Patients Undergoing Radiation Therapy: Opportunities and Challenges. Ann Surg Oncol. Apr;27(4):963–965. DOI: 10.1245/s10434-019-08172-3. PMID: 31898099
Nava MB, Benson JR, Audretsch W, Blondeel P, Catanuto G, Clemens MW, Cordeiro PG, De Vita R, Hammond DC, Jassem J, Lozza L, Orecchia R, Pusic AL, Rancati A, Rezai M, Scaperrotta G, Spano A, Winters ZE, Rocco N(2019) International multidisciplinary expert panel consensus on breast reconstruction and radiotherapy. Br J Surg. Sep;106(10):1327–1340. DOI: 10.1002/bjs.11256. PMID: 31318456
Kronowitz SJ (2012) Current status of implant-based breast reconstruction in patients receiving postmastectomy radiation therapy. Plast Reconstr Surg Oct 130(4):513e–523e. DOI: 10.1097/PRS.0b013e318262f059. PMID: 23018711
Kronowitz SJ (2012) Current status of autologous tissue-based breast reconstruction in patients receiving postmastectomy radiation therapy. Plast Reconstr Surg Aug 130(2):282–292. doi: 10.1097/PRS.0b013e3182589be1. PMID: 22842406
Barry M, Kell MR(2011) Radiotherapy and breast reconstruction: a meta-analysis. Breast Cancer Res Treat. May;127(1):15–22. DOI: 10.1007/s10549-011-1401-x. PMID: 21336948
Tran NV, Chang DW, Gupta A, Kroll SS, Robb GL(2001) Comparison of immediate and delayed free TRAM flap breast reconstruction in patients receiving postmastectomy radiation therapy. Plast Reconstr Surg. Jul;108(1):78–82. DOI: 10.1097/00006534-200107000-00013. PMID: 11420508
Naoum GE, Salama L, Niemierko A, Vieira BL, Belkacemi Y, Colwell AS, Winograd J, Smith B, Ho A, Taghian AG(2020) Single Stage Direct-to-Implant Breast Reconstruction Has Lower Complication Rates Than Tissue Expander and Implant and Comparable Rates to Autologous Reconstruction in Patients Receiving Postmastectomy Radiation. Int J Radiat Oncol Biol Phys. Mar 1;106(3):514–524. DOI: 10.1016/j.ijrobp.2019.11.008. PMID: 31756414
Manyam BV, Shah C, Woody NM, Reddy CA, Weller MA, Juloori A, Naik M, Valente S, Grobmyer S, Durand P, Djohan R, Tendulkar RD(2019) Long-Term Outcomes After Autologous or Tissue Expander/Implant-Based Breast Reconstruction and Postmastectomy Radiation for Breast Cancer. Pract Radiat Oncol. Nov;9(6):e497-e505. DOI: 10.1016/j.prro.2019.06.008. PMID: 31238166
Jagsi R, Momoh AO, Qi J, Hamill JB, Billig J, Kim HM, Pusic AL, Wilkins EG(2018) Impact of Radiotherapy on Complications and Patient-Reported Outcomes After Breast Reconstruction. J Natl Cancer Inst. Feb 1;110(2):157–65. DOI: 10.1093/jnci/djx148. PMID: 28954300
Zhang L, Jin K, Wang X, Yang Z, Wang J, Ma J, Mei X, Chen X, Wang X, Zhou Z, Luo J, Wu J, Shao Z, Zhang Z, Yu X, Guo X(2019) The Impact of Radiotherapy on Reoperation Rates in Patients Undergoing Mastectomy and Breast Reconstruction. Ann Surg Oncol. Apr;26(4):961–968. DOI: 10.1245/s10434-018-07135-4. PMID: 30675702
Reinders FCJ, Young-Afat DA, Batenburg MCT, Bruekers SE, van Amerongen EA, Macaré van Maurik JFM, Braakenburg A, Zonnevylle E, Hoefkens M, Teunis T, Verkooijen HM, van den Bongard HJGD, Maarse W(2020) Higher reconstruction failure and less patient-reported satisfaction after post mastectomy radiotherapy with immediate implant-based breast reconstruction compared to immediate autologous breast reconstruction. Breast Cancer. May;27(3):435–444. DOI 10.1007/s12282-019-01036-4. PMID: 31858435
Rubilar F, Navia A, Dagnino B(2019) Autologous versus prosthetic reconstruction for women with breast cancer who will undergo post-reconstruction radiotherapy. Medwave. Nov 29;19(11):e7728. Spanish, English. DOI: 10.5867/medwave.2019.10.7727. PMID: 31821317
Meresse T, Chaput B, Grolleau JL, Gangloff D(2019) Complication des lambeaux en reconstruction mammaire [Complications of autologous breast reconstruction]. Ann Chir Plast Esthet. Nov;64(5–6):594–619. French. DOI: 10.1016/j.anplas.2019.07.015. PMID: 31383626
Lin AM, Christensen JM, Liao EC, Cetrulo CL Jr, Smith BL, Austen WG Jr, Winograd J, Colwell AS(2021) Postmastectomy Radiation Therapy on Permanent Implants or Tissue Expanders: Which is Better? Ann Surg. Dec 1;274(6):e974-e979. DOI: 10.1097/SLA.0000000000003670. PMID: 31714307
Naoum GE, Oladeru OT, Niemierko A, Salama L, Winograd J, Colwell A, Arafat WO, Smith B, Ho A, Taghian AG(2020) Optimal breast reconstruction type for patients treated with neoadjuvant chemotherapy, mastectomy followed by radiation therapy. Breast Cancer Res Treat. Aug;183(1):127–136. DOI: 10.1007/s10549-020-05747-7. PMID: 32607638
Lee KT, Mun GH(2016) Comparison of one-stage vs two-stage prosthesis-based breast reconstruction: a systematic review and meta-analysis. Am J Surg. Aug;212(2):336 – 44. DOI: 10.1016/j.amjsurg.2015.07.015. PMID: 26499053

Download PDF

Version 1

posted

You are reading this latest preprint version

The effect of adjuvant radiotherapy on one- and two-stage prosthetic breast reconstruction and on autologous reconstruction: a multicenter Italian study among 18 Senonetwork Breast Centres

Status:

Version 1

Abstract

Figures

Introduction

Methods

Sample and Data Collection

Reconstructive Procedure And Endpoints

Statistical analysis

Results

Postsurgical Outcomes And Complications

Postmastectomy Radiotherapy As A Risk Factor For Key Outcomes And Complications

Risk Of Negative Outcomes And Complications By Type Of Procedure

Discussion

Conclusion

Declarations

References

Status:

Version 1