Abstract
Background:
Breast cancer is the most common treatment-related second malignancy among women with previous chest radiotherapy for Hodgkin lymphoma (HL). Little is known about the effects of this kind of radiotherapy on the outcomes of postmastectomy breast reconstruction (BR). This study compared adverse outcomes of BR after HL-related chest radiotherapy to matched controls.
Methods:
The authors conducted a retrospective, matched cohort study in two expert cancer centres in the Netherlands. BRs after therapeutic or prophylactic mastectomy in HL survivors who received chest radiotherapy were matched with BRs in nonirradiated patients without HL on age at mastectomy date, date of BR, and type of BR. The primary outcome was complication-related BR failure or conversion and secondary outcomes were complication-related re-operation, capsular contracture, major donor-site complications, and complication-related ICU admission. The authors analyzed all outcomes univariably using Fisher’s exact tests and the authors assessed reconstruction failure, complication-related re-operation, and capsular contracture with multivariable Cox regression analysis adjusting for confounding and data clustering.
Results:
Seventy BRs in 41 patients who received chest radiotherapy for HL were matched to 121 BRs in 110 nonirradiated patients. Reconstruction failure did not differ between HL survivors (12.9%) and controls (12.4%). The comparison groups showed no differences in number of reoperations, major donor-site complications, or capsular contractures. BR in HL survivors more often let to ICU admission due to complications compared with controls (P=0.048).
Conclusions:
We observed no increased risk of adverse outcomes following BR after previous chest radiotherapy for HL. This is important information for counselling these patients and may improve shared decision-making.
Keywords: breast reconstruction, complications, Hodgkin lymphoma, outcomes, radiotherapy, reconstructive failure
Introduction
Highlights
No increased risk of adverse surgical outcomes was demonstrated following postmastectomy breast reconstruction after previous chest radiotherapy for Hodgkin lymphoma (HL)
Sub-analysis per breast reconstruction technique did not show worse surgical outcomes for HL survivors
Patients with a history of chest radiotherapy for HL may have an increased risk of admission to an ICU after major surgery.
HL survivors currently have a good life expectancy due to effective treatment regimens, including chemotherapy and/or radiotherapy. From the 1960s to the late 1980s, primary treatment for HL frequently comprised extended field radiotherapy such as mantle field radiotherapy (MFRT). MFRT encompassed the supradiaphragmatic lymph node stations, in particular the cervical, supraclavicular, infraclavicular, axillary, hilar, and mediastinal nodes (Fig. 1). Although these radiation protocols proved successful treatments, they are also associated with increased risks of cardiovascular disease, impaired pulmonary or thyroid function and, second malignancies with substantial morbidity and mortality1–12.
Figure 1.
Schematic representation of target area for mantle field irradiation.
Breast cancer (BC) is the most common treatment-related second malignancy among women who received chest radiotherapy for HL, accounting for almost 40% of second cancers4,7,8,11,13. Risks are higher if survivors were irradiated at young ages or received higher radiation doses, with more extensive irradiation fields including the breast3,4,13,14. The increased risk of developing BC is observed from about 10 to at least 40 years after irradiation for HL8,15,16. The risk of BC is increased 37-fold compared with the general population and the cumulative risk of BC may amount to 48% forty years after treatment for HL13,16–20.
BC treatment frequently includes mastectomy21. Postmastectomy breast reconstruction (BR) can improve quality of life and is an important and integrated optional part of BC treatment22. BR may be immediate or delayed, using implants, autologous tissue or a combination23. However, BR may be impeded by radiotherapy for BC due to its negative effects on the chest tissues, which are more profound in implant-based BR than in autologous BR. Therefore, the latter is often preferred after radiotherapy for BC24–26. However, the impact of radiotherapy for HL might differ from that for BC as the two differ in two key respects. First, in BC radiotherapy the dose is higher, with an equivalent dose of 50–66 Gy in 2 Gy fractions, compared with 35–40 Gy in 1.75–2.0 Gy fractions in MFRT. Second, the target area for MFRT does not focus on the breast and the central part of the breast is typically positioned under lung shielding blocks. Studies on the effects of previous radiotherapy for HL on the outcomes of BR are scarce, have small sample sizes, lack appropriate control groups, and show conflicting results27–30. Therefore, the aim of the present study was to compare outcomes after BR in patients with previous chest radiotherapy for HL to matched controls without HL.
Material and methods
We performed a retrospective, matched cohort study nested within a cohort of 3905 HL survivors who were treated in the Netherlands between 1965 and 2000. The selection and methods of data collection for this cohort have been described previously1,8,15,31,32. From this cohort, all women were selected who had been diagnosed with BC after previous chest radiotherapy for HL and had undergone BR after therapeutic or prophylactic mastectomy at the Erasmus MC Cancer Institute in Rotterdam or the Antoni van Leeuwenhoek Hospital in Amsterdam.
BRs of these patients were matched with BRs after therapeutic or prophylactic mastectomy in patients without a history of HL or chest radiotherapy (matched controls). Matching was performed on age at mastectomy (±5 years), date of BR (±5 years), and type of BR (direct-to-implant; tissue expander/implant; autologous with implant; autologous only) and timing of BR (immediate or delayed). We aimed to match each case to two controls. Breasts which had received radiotherapy for BC before BR were excluded. Patients were excluded if their electronic health records had missing data regarding supradiaphragmatic radiotherapy. Four types of BR were included, that is autologous BR, implant-based BR completed in one operation (direct-to-implant) or as a two-stage procedure using a tissue expander and implant subsequently, and BR using a combination of an implant and autologous tissue33.
This study is in accordance with the Declaration of Helsinki, was approved by the Medical Ethical Research Committee of Erasmus MC and is registered at DRKS (protocol accessible)34. This work is in line with the STROCSS criteria35.
Data collection
Data on patient demographics, HL characteristics and treatment, BC characteristics and treatment, BR characteristics, and all reconstruction-related surgical procedures were collected directly from hard copy or electronic patient files.
Outcomes
Primary outcome was BR failure due to complications (referred to as ‘reconstruction failure’), defined as complication-related removal of the reconstruction resulting in a flat chest or conversion to another BR type due to complications.
Secondary outcomes were reoperation due to complications, capsular contracture, major donor-site complications, and complications leading to ICU admission (Clavien-Dindo grade IV)36. Major complications were defined as those requiring reoperation or ICU admission.
Statistical analysis
Baseline characteristics of BRs were compared using frequencies for categorical variables and medians with interquartile ranges for continuous variables. Fisher’s exact tests were used to compare matching variables between cohorts and for univariable analysis of the number of events per outcome measure. If radiotherapy was administered after reconstruction, BRs were censored from the starting date of radiotherapy to prevent obscuring the results with the known effect of radiotherapy for breast cancer. Subgroup analyses was performed for four types of BR.
Differences in reconstruction failure, reoperation due to complications, and capsular contracture between BRs in patients with and without a history chest radiotherapy for HL were assessed using Cox regression models, taking clustering of reconstructions within patients into account. For the first outcome, BRs were censored if a failure occurred due to another reason than a complication (e.g. due to patient dissatisfaction with outcome). For the second and third outcomes, BRs were censored if a failure occurred for any reason.
We considered the following variables as potential confounders: chemotherapy, smoking, ASA-class (American Society of Anesthesiologists Physical Status Classification System)37. Body mass index, age at BR, and bilateral BR. Each potential confounder was added to the base model to assess its influence on the effect of history of HL the depended variable. A variable was retained in the model if the hazard ratios (HR) for history of HL changed by more than 10%. The proportional hazards assumptions were assessed using residual-based methods. Kaplan–Meier curves were plotted to visualize the survival probabilities over time.
Two-sided P values less than 0.05 were considered statistically significant. Analyses were conducted using R-software (version 3.6.1) and the ‘survival’ package (version 3.2-7)38,39.
Results
Baseline comparison of cohorts
Seventy BRs in 41 patients with previous chest radiotherapy for HL were matched to 121 control BRs in 110 patients without a history of HL and without chest radiotherapy. Median age at mastectomy was 47 years (interquartile range=38–52) for HL survivors and 46 years (interquartile range=40–51) for controls. The majority of reconstructions was performed between 2005 and 2015. Immediate reconstruction was performed in 82%. Most reconstructions were implant-based with 44% direct-to-implant and 28% tissue expander/implant procedures. In 9% of reconstructions an implant (±tissue expander) was combined with autologous tissue (latissimus dorsi or thoracodorsal flap). In all alloplastic BRs, the implants were placed in a submuscular position and no acellular dermal matrices were used. Autologous reconstructions were performed in 19% and abdominal tissue was used in 89% of these cases. All BRs in HL survivors were performed following mastectomy for at least unilateral BC compared with 75% in matched controls. This largely explains the difference in the proportion of reconstructions exposed to chemotherapy for BC (41% in HL survivors vs. 25% in controls). HL survivors had more severe comorbidities with 20% being rated ASA III compared with 1% of controls. Smoking habits also differed with more current smokers and less former smokers in the comparison cohort (Table 1).
Table 1.
Baseline characteristics of breast reconstructions performed in Hodgkin lymphoma survivors and matched controls.
| Overall | Breast reconstructions in HL survivors | Controls | P | |
|---|---|---|---|---|
| No. breasts | 191 | 70 | 121 | |
| No. patients | 151 | 41 | 110 | |
| Age at mastectomy, n (%)a | ||||
| <35 years | 24 (12.6) | 10 (14.3) | 14 (11.6) | 0.918 |
| 35–45 years | 59 (30.9) | 22 (31.4) | 37 (30.6) | |
| 45–55 years | 87 (45.5) | 30 (42.9) | 57 (47.1) | |
| >55 years | 21 (11.0) | 8 (11.4) | 13 (10.7) | |
| Year of breast reconstruction, n (%)a | ||||
| <2000 | 19 (9.9) | 8 (11.4) | 11 (9.1) | 0.967 |
| 2000–2005 | 32 (16.8) | 11 (15.7) | 21 (17.4) | |
| 2005–2010 | 73 (38.2) | 26 (37.1) | 47 (38.8) | |
| 2010–2015 | 46 (24.1) | 18 (25.7) | 28 (23.1) | |
| >2015 | 21 (11.0) | 7 (10.0) | 14 (11.6) | |
| Timing of breast reconstuction, n (%)a | ||||
| Immediate | 157 (82.2) | 56 (80.0) | 101 (83.5) | 0.561 |
| Delayed | 34 (17.8) | 14 (20.0) | 20 (16.5) | |
| Type of breast reconstuction, n (%)a | ||||
| Direct-to-implant | 84 (44.0) | 29 (41.4) | 55 (45.5) | 0.854 |
| Tissue expander/implant | 53 (27.7) | 19 (27.1) | 34 (28.1) | |
| Autologousb + implant | 18 (9.4) | 8 (11.4) | 10 (8.3) | |
| Autologousc | 36 (18.8) | 14 (20.0) | 22 (18.2) | |
| Age at breast reconstruction, n (%) | ||||
| <35 years | 20 (10.5) | 9 (12.9) | 11 (9.1) | |
| 35–45 years | 59 (30.9) | 20 (28.6) | 39 (32.2) | |
| 45–55 years | 89 (46.6) | 32 (45.7) | 57 (47.1) | |
| >55 years | 23 (12.0) | 9 (12.9) | 14 (11.6) | |
| Breast cancer, n (%) | ||||
| No | 30 (15.7) | 0 (0.0) | 30 (24.8) | |
| Unilateral | 114 (59.7) | 41 (58.6) | 73 (60.3) | |
| Bilateral | 47 (24.6) | 29 (41.4) | 18 (14.9) | |
| Indication for mastectomy, n (%) | ||||
| Prophylactic | 69 (36.1) | 17 (24.3) | 52 (43.0) | |
| Therapeutic | 122 (63.9) | 53 (75.7) | 69 (57.0) | |
| Mastectomy type, n (%) | ||||
| Skin-sparing mastectomy | 148 (77.5) | 54 (77.1) | 94 (77.7) | |
| Nipple sparing mastectomy | 9 (4.7) | 3 (4.3) | 6 (5.0) | |
| Modified radical mastectomy | 19 (9.9) | 9 (12.9) | 10 (8.3) | |
| Unknown | 15 (7.9) | 4 (5.7) | 11 (9.1) | |
| Laterality reconstructive procedure, n (%) | ||||
| Unilateral procedure | 77 (40.3) | 28 (40.0) | 49 (40.5) | |
| Bilateral procedure | 114 (59.7) | 42 (60.0) | 72 (59.5) | |
| Type of autologous breast reconstuction, n (%) | ||||
| Free flap | 32 (59.3) | 12 (54.5) | 20 (62.5) | |
| Pedicled flap | 22 (40.7) | 10 (45.5) | 12 (37.5) | |
| Mastectomy specimen weight (gram), median [IQR] | 435 [300, 603] | 419 [291, 546] | 449 [321, 630] | |
| Chemotherapy for breast cancer, n (%) | ||||
| Before BR | 25 (13.1) | 13 (18.6) | 12 (9.9) | |
| After BR | 34 (17.8) | 16 (22.9) | 18 (14.9) | |
| No | 132 (69.1) | 41 (58.6) | 91 (75.2) | |
| Endocrine therapy for breast cancer, n (%) | ||||
| Yes | 31 (49.2) | 11 (37.9) | 20 (58.8) | |
| ASA classification, n (%) | ||||
| ASA I | 47 (24.6) | 9 (12.9) | 38 (31.4) | |
| ASA II | 129 (67.5) | 47 (67.1) | 82 (67.8) | |
| ASA III | 15 (7.9) | 14 (20.0) | 1 (0.8) | |
| Body mass index (kg/m2), n (%) | ||||
| <25 | 118 (61.8) | 43 (61.4) | 75 (62.0) | |
| 25–30 | 56 (29.3) | 22 (31.4) | 34 (28.1) | |
| >30 | 17 (8.9) | 5 (7.1) | 12 (9.9) | |
| Smoker, n (%) | ||||
| No | 143 (74.9) | 51 (72.9) | 92 (76.0) | |
| Former | 37 (19.4) | 16 (22.9) | 21 (17.4) | |
| Current | 11 (5.8) | 3 (4.3) | 8 (6.6) | |
| Follow-up (years), median [IQR] | 7.55 [3.46, 10.98] | 8.10 [4.28, 10.86] | 7.04 [3.12, 10.98] | |
ASA, American Society of Anesthesiologists ; BR, breast reconstruction ; HL, Hodgkin lymphoma; IQR, interquartile range.
Variables used for matching.
10 pedicled latissimus dorsi flaps, 8 thoracodorsal flaps.
25 deep inferior epigastric perforator flaps, 1 muscle sparing transverse rectus abdominis muscle (TRAM) flap, 4 free TRAM flaps, 2 pedicled TRAM flaps, 1 superior gluteal artery perforator flap, 1 profunda artery perforator flap, 2 pedicled latissimus dorsi flaps.
The median age at start of irradiation for HL was 21.6 years. The sites of all except two reconstructions in one HL survivor were exposed to Mantle field type radiotherapy. This patient received radiotherapy only to the left cervical and left hilar nodes. Additional therapy for HL included chemotherapy (46% of reconstructions) and splenectomy (20% of reconstructions) (Table 2).
Table 2.
Disease characteristics of Hodgkin lymphoma survivors according to breast reconstruction.
| Breast reconstructions in HL survivors | |
|---|---|
| No. breasts | 70 |
| No. patients | 41 |
| Age at start of irradiation for HL, median [IQR] | 21.61 [18.63, 25.96] |
| HL RTx field, n (%) | |
| Subtotal nodal irradiation | 2 (2.9) |
| Mantle field + para-aortic lymph nodes | 19 (27.1) |
| Mantle field | 44 (62.9) |
| Mantle field excluding axilary nodes | 3 (4.3) |
| Left cervical and hilus nodes | 2 (2.9) |
| Total RTx dosis, median [IQR] | 39.48 [36.30, 40.02] |
| CTx for HL, n (%) | 32 (45.7) |
| Splenectomy, n (%) | 14 (20.0) |
| HL recurrence, n (%) | 7 (10.0) |
CTx, chemotherapy; HL, Hodgkin lymphoma; IQR, interquartile range; RTx, radiotherapy.
Outcomes: univariable analysis
Thirteen percent of all reconstructions failed and this did not differ between cohorts (12.9% in HL survivors vs. 12.4% in controls). The percentage of reconstructions which did not need any complication-related revision surgery was similar for reconstructions in HL survivors compared with matched controls (66% vs. 64%, respectively). However, if a complication occurred that required surgery, HL survivors underwent more procedures compared with matched controls. This difference was statistically significant for tissue expander/implant BRs (P=0.001). In three HL survivors, complications after autologous reconstruction resulted in ICU admissions while this did not occur in the comparison cohort (P=0.048) (Table 3). One patient suffered from postoperative hypoxia due to atelectasis after DIEP flap reconstruction, one patient developed re-entry tachycardia for which she needed treatment with adenosine after latissimus dorsi flap reconstruction with a tissue expander, and one patient required prolonged ICU admission due to idiopathic low oxygen saturation after revision surgery for total flap failure of her SGAP flap reconstruction. All patients were ASA II.
Table 3.
Adverse outcomes of breast reconstructions in Hodgkin lymphoma survivors compared with matched controls: all reconstruction types combined.
| Breast reconstructions in HL survivors | Controls | P | |
|---|---|---|---|
| n | 70 | 121 | |
| Failure or conversion of breast reconstruction due to a complication, n (%) | 9 (12.9) | 15 (12.4) | 1.000 |
| Reoperations due to a complication, n (%) | |||
| 0 | 46 (65.7) | 77 (63.6) | 0.350 |
| 1 | 14 (20.0) | 33 (27.3) | |
| 2+ | 10 (14.3) | 11 (9.1) | |
| ICU Admission, n (%) | 3 (4.3) | 0 (0.0) | 0.048a |
HL, Hodgkin lymphoma.
Indicates a significant test result.
Sub-analysis per BR type showed no differences between HL survivors and matched controls in reconstruction failure, reoperation rates, capsular contracture or major donor-site complications, except for reoperation rates after tissue expander/implant BR (Table 4). When combining all implant-related reconstructions, capsular contracture occurred less frequently in HL survivors than in the comparison cohort (5.4% vs. 17.2% respectively, P=0.045). There was no significant difference in major donor-site complications when all reconstructions involving autologous tissue were combined (HL cohort 9.1% vs. comparison cohort 12.5%, P=1.00).
Table 4.
Adverse outcomes of breast reconstructions in Hodgkin lymphoma survivors compared with matched controls: stratified per reconstruction type.
| Breast reconstructions in HL survivors | Controls | P | |
|---|---|---|---|
| Direct-to-implant | |||
| n | 29 | 55 | |
| Failure or conversion of breast reconstruction due to a complication, n (%) | 5 (17.2) | 8 (14.5) | 0.759 |
| Reoperations due to a complication, n (%) | |||
| 0 | 20 (69.0) | 36 (65.5) | 1.000 |
| 1 | 6 (20.7) | 12 (21.8) | |
| 2+ | 3 (10.3) | 7 (12.7) | |
| Capsular contracture, n (%) | 1 (3.4) | 8 (14.5) | 0.154 |
| Tissue expander/implant | |||
| n | 19 | 34 | |
| Failure or conversion of breast reconstruction due to a complication, n (%) | 3 (15.8) | 5 (14.7) | 1.000 |
| Reoperations due to a complication, n (%) | |||
| 0 | 12 (63.2) | 18 (52.9) | 0.001a |
| 1 | 2 (10.5) | 16 (47.1) | |
| 2+ | 5 (26.3) | 0 (0.0) | |
| Capsular contracture, n (%) | 2 (10.5) | 9 (26.5) | 0.290 |
| Autologous with implant | |||
| n | 8 | 10 | |
| Failure or conversion of breast reconstruction due to a complication, n (%) | 0 (0.0) | 2 (20.0) | 0.477 |
| Reoperations due to a complication, n (%) | |||
| 0 | 7 (87.5) | 6 (60.0) | 0.588 |
| 1 | 1 (12.5) | 2 (20.0) | |
| 2+ | 0 (0.0) | 2 (20.0) | |
| Capsular contracture, n (%) | 0 (0.0) | 0 (0.0) | 1.000 |
| Major donor-site complications, n (%) | 0 (0.0) | 0 (0.0) | 1.000 |
| Autologous only | |||
| n | 14 | 22 | |
| Failure or conversion of breast reconstruction due to a complication, n (%) | 1 (7.1) | 0 (0.0) | 0.389 |
| Reoperations due to a complication, n (%) | |||
| 0 | 7 (50.0) | 17 (77.3) | 0.200 |
| 1 | 5 (35.7) | 3 (13.6) | |
| 2+ | 2 (14.3) | 2 (9.1) | |
| Major donor-site complications, n (%) | 2 (14.3) | 4 (18.2) | 1.000 |
HL, Hodgkin lymphoma.
Indicates a significant test result.
Outcomes: multivariable analysis
The HR for reconstruction failure comparing reconstructions in HL survivors to matched controls was 0.81 (95% CI: 0.34–1.96) in multivariable analysis adjusting for post-reconstruction chemotherapy (Table 5). Post-reconstruction chemotherapy was a risk factor for reconstruction failure (HR: 7.6; 95% CI: 3.4–16.9; P<0.001). Figure 2 shows the univariable Kaplan–Meier survival curves for reconstructions in HL survivors and matched controls.
Table 5.
Risks of various complications comparing breast reconstructions in Hodgkin lymphoma survivors with matched control, based on Cox regression models.
| Univariable | Multivariable | |||||
|---|---|---|---|---|---|---|
| HR | 95% CI | P | HR | 95% CI | P | |
| Failure or conversion due to a complication | 1.03 | 0.41–2.61 | 0.95 | 0.81a | 0.34–1.96 | 0.64 |
| Reoperation due to a complication | 0.93 | 0.53–1.61 | 0.79 | 0.75b | 0.40–1.43 | 0.39 |
| Capsular contracture | 0.28 | 0.07–1.15 | 0.08 | 0.33c | 0.08–1.39 | 0.13 |
HR, hazard ratio.
Corrected for post-reconstruction chemotherapy.
Corrected for ASA-class.
Corrected for active smoking.
Both the univariable and multivariable model take clustering per patient into account.
Figure 2.
Univariable survival curve (Kaplan–Meier) for survival until failure or conversion of breast reconstruction due to a complication (unadjusted for clustering/confounding). HL, Hodgkin lymphoma.
The HR for complication-related reoperations comparing reconstructions in HL survivors to controls was 0.75 (95% CI: 0.40–1.43) in multivariable analysis adjusted for ASA-class (Table 5). Figure 3 shows univariable time-to-event plots (Kaplan–Meier) for complication-related reoperations in HL survivors and matched controls.
Figure 3.
Univariable survival curve (Kaplan–Meier) for reoperation due to a complication after breast reconstruction (unadjusted for clustering/confounding). HL, Hodgkin lymphoma.
The HR for capsular contracture comparing reconstructions in HL survivors to controls was 0.33 (95% CI: 0.08–1.39) in multivariable analysis adjusted for active smoking (Table 5). Active smoking was a risk factor for this outcome (HR: 5.01; 95% CI: 1.66–15.11; P=0.004). Figure 4 shows univariable time-to-event plots (Kaplan–Meier) for capsular contracture for all implant-based reconstructions in HL survivors and matched controls.
Figure 4.
Univariable survival curve (Kaplan–Meier) for survival until capsular contracture for all breast reconstructions involving an implant (unadjusted for clustering/confounding). HL, Hodgkin lymphoma.
Salvage BR
Thirteen salvage BRs were performed after the initial reconstruction failed due to complications; four in HL survivors and nine in the controls. These consisted of four tissue expander/implant BRs, three reconstructions combining an implant and autologous tissue, four microvascular autologous reconstructions, and two only using lipofilling. All salvage BRs were surgically successful.
Discussion
Offering BR to HL survivors with previous chest radiotherapy for HL remains controversial because of presumed increased risks of adverse outcomes. The present study represents the largest series of BR in HL survivors and is the first study to compare outcomes with matched controls. Our study showed no increased risks for developing complications leading to BR failure or conversion, reoperation, major donor-site complications, or capsular contracture in HL survivors.
BR in HL survivors resulted in more admissions to the ICU than in the comparison cohort. Although these HL survivors were categorized as ASA II, they needed (prolonged) ICU admission due to pulmonary and cardiac problems. This may be explained by the known increased risk of cardiovascular and pulmonary disease which might not have been recognized preoperatively resulting in a low ASA classification1,5,10. Notably, these ICU admissions were only seen after autologous flap reconstruction (both pedicled and free) which might be associated with a longer operation time in flap-based reconstruction compared with implant-based reconstruction. Our small sample size prevents a strong recommendation, but it seems advisable to preoperatively discuss the risk of postoperative complications in HL survivors associated with their history of radiotherapy for HL.
Sub-analysis per reconstruction type showed no differences between the matched cohorts except in reoperation rates after tissue expander/implant BR. Less tissue expander/implant BRs in HL survivors needed any revision surgery due to complications compared with controls. However, if any reoperation was required, generally more procedures were necessary to solve the complication. Another remarkable result was the lower capsular contracture rate in HL survivors compared with controls when analyzing all implant-related reconstructions.
Remarkably, chemotherapy after completing BR was an independent predictor of reconstruction failure compared with no or pre-reconstruction chemotherapy. It mostly concerned immediate implant-based BRs, which were exposed to adjuvant chemotherapy. Post-reconstruction chemotherapy, compared with no or preoperative chemotherapy, may be associated with increased risk of complications and failure due to the negative effect of chemotherapy on wound healing40–42. If chemotherapy is planned before BR, it is possible to postpone BR in case a patient experiences a significant chemotherapy-related health deterioration, thereby avoiding the development of complications while weaker. However, if chemotherapy is planned after BR, any delay in the start of chemotherapy usually is minimalized, risking commencement while postoperative complications may still exist. Previous literature has reported conflicting results on this topic43,44.
Active smoking was an independent predictor for developing capsular contracture. This association was previously also reported by some studies, but could not be found by others45–47. Our finding might be explained by our longer follow-up compared with earlier literature. It may be an incidental finding, especially given our small cohort size. However, smoking is known to cause a systemic inflammatory response and is associated with decreased tissue oxygenation and increased risk of infection48. It is conceivable that smoking increases the risk of capsular contracture, because a popular hypothesis for the pathogenesis of capsular contracture involves chronic infection and inflammation49. In the present study, BRs in HL survivors were less often exposed to smoking than matched controls.
Radiotherapy for BC is associated with a higher risk of complications after BR24 Implant-based reconstruction has been shown to be associated with poorer outcomes than autologous reconstruction24–26. Our study suggest that BR can be safely done in patients with a history of chest radiotherapy for HL. This can be explained partly by the lower average dose of 39 Gy in HL survivors in our study compared with 50 Gy in BC patients, which is even higher if a boost is given. Also, during MFRT a proportion of the breast and overlying skin is protected with lung shielding blocks, protecting the future reconstruction site. Late effects of radiotherapy such as fibrosis may progress many years after treatment which could be a disadvantage for HL survivors as their interval between radiotherapy for HL and reconstruction is usually considerably longer than a typical interval between radiotherapy for BC and reconstruction50. However, this longer interval also means that the severity of the late effects is clear by the time of reconstruction and can be taken into consideration when deciding on the reconstructive approach. Based on literature regarding the effect of radiotherapy for BC, one could argue autologous BR may be preferred over implant-based BR in HL survivors, however, our results showed no need for such a preference.
All salvage reconstructions after initially failed BR were surgically successful. This offers some perspective to those patients who initially experience a BR failure.
Only four studies on BR after chest radiotherapy for HL have been published previously, all of which lacked a comparison group of nonirradiated patients27–30. Bacilious et al. 28 were the first to publish a series of seven patients who underwent eleven immediate tissue expander/implant BRs after MFRT for HL; they reported only one complication (9%) and no BR failure after a mean follow-up of 3 years Wong et al.27 evaluated the postoperative outcomes of 23 implant-based and four autologous reconstructions after mastectomy in 16 HL survivors who had received MFRT. They reported a higher overall complication rate of 69% after an average follow-up of 3.7 years compared with Bacillious et al. 27,28 Five implant-based reconstructions required implant removal and conversion to autologous reconstructions (22%), which is comparable to our 17%-rate after 7.6 years median follow-up in implant-based reconstruction without autologous tissue. The difference in overall complications may partly be explained because their series contained only 4% delayed reconstructions compared with 18% in our group. Freniere et al.29 evaluated 79 immediate implant-based and 18 immediate autologous reconstructions in 97 HL survivors who had received MFRT with a follow-up of 5.6 years. They reported a complication rate of 33% and an unplanned operative revision rate of 30% overall, 21% for implant-based reconstruction and 72% for autologous reconstruction. We did not observe these large differences between BR types. They explained the relatively high proportion of unplanned revisions in their autologous group by hypothesizing a potential lower threshold for revisions in the healthy, well vascularized tissue of an autologous BR compared with revising scars over implants in an irradiated mastectomy skin envelope. For the same reason, one of the two expert centres in our study preferred autologous over implant-based BR. Freniere et a. reported a complete reconstruction failure rate of 3.7% for implant-based and 0% for autologous reconstruction, averaging 3.0% overall. We reported 12.9% amongst HL survivors, but this included 2 conversions, leaving 10% (n=7) when restricted to failure only. Van Huizum et al.30 compared 42 immediate implant-based breast reconstructions after skin-sparing mastectomy in patients receiving radiotherapy for HL to 47 salvage breast reconstruction after breast conserving therapy and associated radiotherapy for BC. They reported an increased risk for adverse outcomes after radiotherapy for BC, but not after radiotherapy for HL.
Our overall BR failure rates compare unfavourable to literature in general. For implant-based BR we observed a 15.3% failure rate compared with 3.3–5.4% in recent meta-analyses51–53. Another meta-analysis reported data separately for direct-to-implant (14.4%) and two-stage reconstruction (8.7%) techniques, compared with 15.5% and 15.1%, respectively, in the present study54. Massenburg et al.55 reported 1.1% flap failure for a cohort latissimus dorsi flap reconstructions combined with an implant which is notebly lower than our 11.1% failure rate (including conversions and implant-related failures) in the autologous plus implant cohort containing both latissimus dorsi and thoracodorsal artery perforator flaps. Finally, we had 2.8% reconstruction failures in the autologous cohort which is comparable to rates reported in the literature of 1.0–4.4%53,56–58. It is not fully clear why we found relatively high failure rates. Partly, it may be explained by a referral bias as patients were selected from two tertiary medical centres who generally operate more complex patients. More importantly, our definition of reconstruction failure also included conversions to another reconstruction type. This broad definition also considered direct-to-implant reconstructions as failed if they required a tissue expander at any point, and reconstructions based on implants only were also scored as failed if a small thoracodorsal artery perforator flap was added secondarily. Finally, our small sample sizes may have attributed to inaccurate estimates.
Although the present study reports the first comparison of BRs in HL survivors who received chest radiotherapy with matched controls, it is limited by its retrospective study design and small sample size. We chose to focus on objective outcomes involving reoperation because reliable data on such outcomes can be collected retrospectively. Although selection bias due to loss-to-follow-up might have occurred as patients with more severe complications tend to be followed longer, our long-term follow-up of 7.6 years exceeds that of earlier publications. Furthermore, we adjusted for a number of potential confounding factors and data clustering. The limited number of outcome events reduced statistical power when making more detailed comparisons by reconstruction technique.
Conclusion
This first comparative study evaluating BR outcomes in HL survivors provides reliable and reassuring data suggesting that there is no increased risk of adverse surgical outcomes following BR after previous chest radiotherapy for HL, either using implants or autologous tissue. This is important information for counselling HL survivors with BC who need to undergo mastectomy and consider breast reconstruction and may also improve shared decision-making. In addition, our findings may be extrapolated to BR for breast cancer developed after thoracic radiotherapy exposures for other childhood or young adult cancers, such as sarcoma and thymoma where the breast receives an incidental dose.
Ethical approval
The MERC of the Erasmus Medical Center, Rotterdam, The Netherlands, judged our research protocol positively (MEC-2022-0258).
Source of funding
No funding was received for this article.
Author contribution
All the authors have made a substantial contribution to the concept or design of the article, or the acquisition, analysis or interpretation of data for the article, and to drafting the work or revising it critically for important intellectual content.
Conflicts of interest disclosure
The Department of Plastic & Reconstructive Surgery of the Erasmus MC received funding from POLYTECH Health & Aesthetics GmbH in the form of financial support for personnel costs for one PhD candidate (no grant number available) from July 2018 until July 2021. No funding was received for this article. None of the authors has a financial interest in any of the products, devices, or drugs mentioned in this manuscript.
Research registration unique identifying number (UIN)
Name of the registry: Deutschen Register Klinischer Studien (DRKS).
Unique Identifying number or registration ID: DRKS00028655.
Hyperlink to your specific registration https://drks.de/search/en/trial/DRKS00028655.
Guarantor
JX Harmeling and MAM Mureau.
Provenance and peer review
Not commissioned, externally peer-reviewed.
Footnotes
Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article.
Published online 11 April 2023
Contributor Information
J. Xavier Harmeling, Email: xavier.harmeling@gmail.com.
Leonie A.E. Woerdeman, Email: l.woerdeman@nki.nl.
Ezgi Ozdemir, Email: ozezgi@hotmail.com.
Michael Schaapveld, Email: m.schaapveld@nki.nl.
Hester S.A. Oldenburg, Email: h.oldenburg@nki.nl.
Cécile P.M. Janus, Email: c.janus@erasmusmc.nl.
Nicola S. Russell, Email: n.russell@nki.nl.
Linetta B. Koppert, Email: l.koppert@erasmusmc.nl.
Inge M. Krul, Email: i.krul@nki.nl.
Flora E. van Leeuwen, Email: f.v.leeuwen@nki.nl.
Marc A.M. Mureau, Email: m.mureau@erasmusmc.nl.
References
- 1.Aleman BM, van den Belt-Dusebout AW, De Bruin ML, et al. Late cardiotoxicity after treatment for Hodgkin lymphoma. Blood 2007;109:1878–1886. [DOI] [PubMed] [Google Scholar]
- 2.Alm El-Din MA, El-Badawy SA, Taghian AG. Breast cancer after treatment of Hodgkin’s lymphoma: general review. Int J Radiat Oncol Biol Phys 2008;72:1291–1297. [DOI] [PubMed] [Google Scholar]
- 3.Cooke R, Jones ME, Cunningham D, et al. Breast cancer risk following Hodgkin lymphoma radiotherapy in relation to menstrual and reproductive factors. Br J Cancer 2013;108:2399–2406. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Hodgson DC. Late effects in the era of modern therapy for Hodgkin lymphoma. Hematol Am Soc Hematol Educ Program 2011;2011:323–329. [DOI] [PubMed] [Google Scholar]
- 5.Matasar MJ, Ford JS, Riedel ER, et al. Late morbidity and mortality in patients with Hodgkin’s lymphoma treated during adulthood. J Natl Cancer Inst 2015;107:djv018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Mazonakis M, Lyraraki E, Damilakis J. Second cancer risk assessments after involved-site radiotherapy for mediastinal Hodgkin lymphoma. Med Phys 2017;44:3866–74. [DOI] [PubMed] [Google Scholar]
- 7.Milano MT, Li H, Gail MH, et al. Long-term survival among patients with Hodgkin’s lymphoma who developed breast cancer: a population-based study. J Clin Oncol 2010;28:5088–5096. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Schaapveld M, Aleman BM, van Eggermond AM, et al. second cancer risk up to 40 years after treatment for Hodgkin’s lymphoma. N Engl J Med 2015;373:2499–2511. [DOI] [PubMed] [Google Scholar]
- 9.van Leeuwen FE, Ng AK. Long-term risk of second malignancy and cardiovascular disease after Hodgkin lymphoma treatment. Hematol Am Soc Hematol Educ Program 2016;2016:323–230. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.van Nimwegen FA, Schaapveld M, Janus CP, et al. Cardiovascular disease after Hodgkin lymphoma treatment: 40-year disease risk. JAMA Intern Med 2015;175:1007–1017. [DOI] [PubMed] [Google Scholar]
- 11.Veit-Rubin N, Rapiti E, Usel M, et al. Risk, characteristics, and prognosis of breast cancer after Hodgkin’s lymphoma. Oncologist 2012;17:783–791. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.de Vries S, Schaapveld M, Janus CPM, et al. Long-term cause-specific mortality in Hodgkin lymphoma patients. J Natl Cancer Inst 2021;113:760–769. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Basu SK, Schwartz C, Fisher SG, et al. Unilateral and bilateral breast cancer in women surviving pediatric Hodgkin’s disease. Int J Radiat Oncol Biol Phys 2008;72:34–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Tinger A, Wasserman TH, Klein EE, et al. The incidence of breast cancer following mantle field radiation therapy as a function of dose and technique. Int J Radiat Oncol Biol Phys 1997;37:865–870. [DOI] [PubMed] [Google Scholar]
- 15.De Bruin ML, Sparidans J, van’t Veer MB, et al. Breast cancer risk in female survivors of Hodgkin’s lymphoma: lower risk after smaller radiation volumes. J Clin Oncol 2009;27:4239–4246. [DOI] [PubMed] [Google Scholar]
- 16.Krul IM, Opstal-van Winden AWJ, Aleman BMP, et al. Breast cancer risk after radiation therapy for Hodgkin lymphoma: influence of gonadal hormone exposure. Int J Radiat Oncol Biol Phys 2017;99:843–53. [DOI] [PubMed] [Google Scholar]
- 17.Swerdlow AJ, Cooke R, Bates A, et al. Breast cancer risk after supradiaphragmatic radiotherapy for Hodgkin’s lymphoma in England and Wales: a National Cohort Study. J Clin Oncol 2012;30:2745–2752. [DOI] [PubMed] [Google Scholar]
- 18.Moskowitz CS, Chou JF, Wolden SL, et al. Breast cancer after chest radiation therapy for childhood cancer. J Clin Oncol 2014;32:2217–2223. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Travis LB, Hill D, Dores GM, et al. Cumulative absolute breast cancer risk for young women treated for Hodgkin lymphoma. J Natl Cancer Inst 2005;97:1428–1437. [DOI] [PubMed] [Google Scholar]
- 20.Bhatia S, Robison LL, Oberlin O, et al. Breast cancer and other second neoplasms after childhood Hodgkin’s disease. N Engl J Med 1996;334:745–751. [DOI] [PubMed] [Google Scholar]
- 21.https://dica.nl/jaarrapportage-2018 DICA - Dutch Institute for Clinical Auditing. Jaarraportage 2018 Dutch Breast Implant Registry (DBIR) 2018 Available from: [PubMed]
- 22.Kouwenberg CAE, de Ligt KM, Kranenburg LW, et al. Long-term health-related quality of life after four common surgical treatment options for breast cancer and the effect of complications: a retrospective patient-reported survey among 1871 patients. Plast Reconstr Surg 2020;146:1–13. [DOI] [PubMed] [Google Scholar]
- 23.Mureau MAM. Breast Reconstruction Guideline Working G. Dutch breast reconstruction guideline. J Plast Reconstr Aesthet Surg 2018;71:290–304. [DOI] [PubMed] [Google Scholar]
- 24.Barry M, Kell MR. Radiotherapy and breast reconstruction: a meta-analysis. Breast Cancer Res Treat 2011;127:15–22. [DOI] [PubMed] [Google Scholar]
- 25.Magill LJ, Robertson FP, Jell G, et al. Determining the outcomes of post-mastectomy radiation therapy delivered to the definitive implant in patients undergoing one- and two-stage implant-based breast reconstruction: a systematic review and meta-analysis. J Plast Reconstr Aesthet Surg 2017;70:1329–35. [DOI] [PubMed] [Google Scholar]
- 26.Yun JH, Diaz R, Orman AG. Breast reconstruction and radiation therapy. Cancer Control 2018;25:1073274818795489. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Wong RK, Morrison SD, Momeni A, et al. Outcomes of breast reconstruction in breast cancer patients with a history of mantle radiation for Hodgkin lymphoma. Ann Plast Surg 2014;72(suppl 1):S46–S50. [DOI] [PubMed] [Google Scholar]
- 28.Bacilious N, Cordeiro PG, Disa JJ, et al. Breast reconstruction using tissue expanders and implants in Hodgkin’s patients with prior mantle irradiation. Plast Reconstr Surg 2002;109:102–107. [DOI] [PubMed] [Google Scholar]
- 29.Freniere B, Irwin T, Guldbrandsen G, et al. Outcomes following breast reconstruction in patients with prior mantle radiation for treatment of Hodgkin’s lymphoma. Ann Plast Surg 2020;84:507–11. [DOI] [PubMed] [Google Scholar]
- 30.van Huizum MA, Hage JJ, Russell NS, et al. Combined skin-sparing mastectomy and immediate implant-based breast reconstruction: Outcome following mantle field irradiation versus outcome following whole-breast irradiation. J Plast Reconstr Aesthet Surg 2022;75:94–103. [DOI] [PubMed] [Google Scholar]
- 31.van Leeuwen FE, Klokman WJ, Hagenbeek A, et al. Second cancer risk following Hodgkin’s disease: a 20-year follow-up study. J Clin Oncol 1994;12:312–325. [DOI] [PubMed] [Google Scholar]
- 32.van Eggermond AM, Schaapveld M, Lugtenburg PJ, et al. Risk of multiple primary malignancies following treatment of Hodgkin lymphoma. Blood 2014;124:319–327; quiz 466. [DOI] [PubMed] [Google Scholar]
- 33.Ahmed S, Snelling A, Bains M, et al. Breast reconstruction. BMJ 2005;330:943–948. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Deutschen Register Klinischer Studien (DRKS). Registration number: DRKS00028655. Available from: https://drks.de/search/en/trial/DRKS00028655
- 35.Mathew G, Agha R. Group S. STROCSS 2021: Strengthening the reporting of cohort, cross-sectional and case-control studies in surgery. Int J Surg 2021;96:106165. [DOI] [PubMed] [Google Scholar]
- 36.Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004;240:205–213. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Hurwitz EE, Simon M, Vinta SR, et al. Adding examples to the ASA-physical status classification improves correct assignment to patients. Anesthesiology 2017;126:614–22. [DOI] [PubMed] [Google Scholar]
- 38.Therneau TM. A Package for Survival Analysis in R. 2020. Available from: https://CRAN.R-project.org/package=survival
- 39.RStudio Team “RStudio I. RStudio: Integrated Development Environment for R.: RStudio, Inc; 2015. 3.6.1. ed. [Google Scholar]
- 40.Dolen UC, Schmidt AC, Um GT, et al. Impact of neoadjuvant and adjuvant chemotherapy on immediate tissue expander breast reconstruction. Ann Surg Oncol 2016;23:2357–2366. [DOI] [PubMed] [Google Scholar]
- 41.Lam TC, Borotkanics R, Hsieh F, et al. Immediate two-stage prosthetic breast reconstruction failure: radiation is not the only culprit. Plast Reconstr Surg 2018;141:1315–24. [DOI] [PubMed] [Google Scholar]
- 42.Warren Peled A, Itakura K, Foster RD, et al. Impact of chemotherapy on postoperative complications after mastectomy and immediate breast reconstruction. Arch Surg 2010;145:880–885. [DOI] [PubMed] [Google Scholar]
- 43.El-Sabawi B, Sosin M, Carey JN, et al. Breast reconstruction and adjuvant therapy: a systematic review of surgical outcomes. J Surg Oncol 2015;112:458–464. [DOI] [PubMed] [Google Scholar]
- 44.Kooijman MML, Hage JJ, Oldenburg HSA, et al. Surgical complications of skin-sparing mastectomy and immediate implant-based breast reconstruction in women concurrently treated with adjuvant chemotherapy for breast cancer. Ann Plast Surg 2021;86:146–50. [DOI] [PubMed] [Google Scholar]
- 45.Bachour Y, Bargon CA, de Blok CJM, et al. Risk factors for developing capsular contracture in women after breast implant surgery: a systematic review of the literature. J Plast Reconstr Aesthet Surg 2018;71:e29–e48. [DOI] [PubMed] [Google Scholar]
- 46.de Kerckhove M, Iwahira Y. Risk factors for capsular contracture: a retrospective study in immediate reconstruction versus delayed reconstruction. Plast Reconstr Surg Glob Open 2020;8:e2864. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Hammond JB, Kosiorek HE, Cronin PA, et al. Capsular contracture in the modern era: A multidisciplinary look at the incidence and risk factors after mastectomy and implant-based breast reconstruction. Am J Surg 2021;221:1005–1010. [DOI] [PubMed] [Google Scholar]
- 48.Sorensen LT. Wound healing and infection in surgery: the pathophysiological impact of smoking, smoking cessation, and nicotine replacement therapy: a systematic review. Ann Surg 2012;255:1069–1079. [DOI] [PubMed] [Google Scholar]
- 49.Bachour Y, Verweij SP, Gibbs S, et al. The aetiopathogenesis of capsular contracture: a systematic review of the literature. J Plast Reconstr Aesthet Surg 2018;71:307–317. [DOI] [PubMed] [Google Scholar]
- 50.Dorr W. Radiobiology of tissue reactions. Ann ICRP 2015;44(1 suppl):58–68. [DOI] [PubMed] [Google Scholar]
- 51.Han SE, Lee KT, Bang S. Comprehensive comparison between shaped versus round implants for breast reconstruction: a systematic review and meta-analysis. Aesthet Surg J 2021;41:34–44. [DOI] [PubMed] [Google Scholar]
- 52.Li Y, Xu G, Yu N, et al. Prepectoral versus subpectoral implant-based breast reconstruction: a meta-analysis. Ann Plast Surg 2020;85:437–47. [DOI] [PubMed] [Google Scholar]
- 53.Tsoi B, Ziolkowski NI, Thoma A, et al. Safety of tissue expander/implant versus autologous abdominal tissue breast reconstruction in postmastectomy breast cancer patients: a systematic review and meta-analysis. Plast Reconstr Surg 2014;133:234–49. [DOI] [PubMed] [Google Scholar]
- 54.Basta MN, Gerety PA, Serletti JM, et al. A systematic review and head-to-head meta-analysis of outcomes following direct-to-implant versus conventional two-stage implant reconstruction. Plast Reconstr Surg 2015;136:1135–44. [DOI] [PubMed] [Google Scholar]
- 55.Massenburg BB, Sanati-Mehrizy P, Ingargiola MJ, et al. Flap failure and wound complications in autologous breast reconstruction: a national perspective. Aesthetic Plast Surg 2015;39:902–909. [DOI] [PubMed] [Google Scholar]
- 56.Cao Z, Cao J, Pang X, et al. A comparative study for the rate of adverse outcomes in unilateral and bilateral abdominal flap breast reconstruction: a meta-analysis. Medicine (Baltimore) 2020;99:e22096. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 57.Khajuria A, Prokopenko M, Greenfield M, et al. A meta-analysis of clinical, patient-reported outcomes and cost of DIEP versus implant-based breast reconstruction. Plast Reconstr Surg Glob Open 2019;7:e2486. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 58.Jeong W, Lee S, Kim J. Meta-analysis of flap perfusion and donor site complications for breast reconstruction using pedicled versus free TRAM and DIEP flaps. Breast 2018;38:45–51. [DOI] [PubMed] [Google Scholar]