European Journal of Cancer (2014) 50, 2752– 2762

Available at www.sciencedirect.com

ScienceDirect journal homepage: www.ejcancer.com

Review

‘Reconstruction: Before or after postmastectomy radiotherapy?’ A systematic review of the literature Judith Berbers a, Angela van Baardwijk b, Ruud Houben b, Esther Heuts c, Marjolein Smidt c, Kristien Keymeulen c, Maud Bessems c,1, Stefania Tuinder d, Liesbeth J. Boersma b,⇑ a

Maastricht University Medical Center, Faculty of Health, Medicine and Life Sciences, Maastricht, The Netherlands Maastricht University Medical Center, Dept. Radiation Oncology (MAASTRO Clinic), GROW, Maastricht, The Netherlands c Maastricht University Medical Center, Dept. Surgical Oncology, Maastricht, The Netherlands d Maastricht University Medical Center, Dept. Reconstructive Surgery, Maastricht, The Netherlands b

Received 27 March 2014; received in revised form 10 July 2014; accepted 30 July 2014 Available online 26 August 2014

KEYWORDS Breast cancer Mastectomy Radiotherapy Breast reconstruction Autologous reconstruction Implant reconstruction Immediate reconstruction Delayed reconstruction Complications Cosmetic outcome

Abstract Objective: The aim of this review is to investigate the effect of timing of the reconstruction and radiotherapy, with respect to complication rate and cosmetic outcome, with a special focus on the timing of the placement of the definite implant. Methods: PubMed was searched for publications between January 2000 and December 2012. Of 37 eligible studies, timing of reconstruction, type, and incidence of complications were recorded. First, we calculated the weighted mean including confidence intervals for complications and cosmetic outcome overall, and for the following subgroups: (1) Autologous reconstruction after radiotherapy; (2) Definite implant reconstruction after radiotherapy; (3) Autologous reconstruction before radiotherapy; (4) Definite implant reconstruction before radiotherapy. A second analysis was performed using only studies that directly compared group 1 versus 3 and 2 versus 4. Results: A large variation in complication rates (8.7–70.0%) and in acceptable cosmetic outcome (41.4–93.3%) was reported. The first analysis showed more complications and a higher revision rate if an implant reconstruction was performed after radiotherapy; for autologous reconstruction fibrosis occurred more often if reconstruction was applied first. The second analysis showed no significant differences in total complication rate. Only implant failure occurred more often if applied after radiotherapy (odds ratio (OR) 3.03 [1.59–5.77]). No differences were found in both patient and physician satisfaction.

⇑ Corresponding author: Address: Dr. Tanslaan 12, 6229 ET Maastricht, The Netherlands. Tel.: +31 88 44 55 666; fax: +31 88 44 55 773. 1

E-mail address: [email protected] (L.J. Boersma). Current address: Jeroen Bosch Hospital, Den Bosch, The Netherlands.

http://dx.doi.org/10.1016/j.ejca.2014.07.023 0959-8049/Ó 2014 Elsevier Ltd. All rights reserved.

J. Berbers et al. / European Journal of Cancer 50 (2014) 2752–2762

2753

Conclusions: A definite implant reconstruction placed before radiotherapy limits the rate of complications. For autologous reconstruction, less fibrosis is seen if reconstruction is performed after radiotherapy, but timing had no significant impact on total complication rate. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction Breast cancer is the most common type of cancer in women and the leading cause of cancer related mortality in women [1]. In general, surgical treatment of breast cancer can be divided into breast-conserving therapy or mastectomy, with equal survival rates in early breast cancer patients [2]. Due to the introduction of magnetic resonance imaging (MRI) [3–5], screening for BRCA-1 and BRCA-2 gene mutations [5,6], in addition to technical surgical improvements, mastectomy followed by reconstruction is increasingly being chosen as a preferred treatment option. In the case of mastectomy with reconstruction, a choice has to be made between autologous reconstruction or implant reconstruction. Both types of reconstructions can be performed immediately after mastectomy or delayed. Patients with immediate reconstruction have reported better quality of life compared to patients with a delayed breast reconstruction [7,8]. However, in some high-risk cases postmastectomy radiotherapy is indicated [9], which increases the complication rate of any reconstruction [10–15]. Hard evidence about the optimal sequence of radiotherapy and reconstruction, for different types of reconstruction is lacking. Most guidelines nowadays however, recommend delaying reconstruction, if it is preoperatively known that post-operative radiotherapy is indicated. However, in some situations, the indication for post-mastectomy radiotherapy only becomes clear after mastectomy. In that case, four situations can be discriminated, with subsequent practical clinical questions:

4) A complete immediate implant reconstruction has been performed: A) What is the effect of radiotherapy on the complication rate and cosmetic outcome? Recently, five other reviews [16–20] on reconstruction and radiotherapy have been published. All these papers only dealt with part of our questions, or different questions. The aim of the current study was therefore, to investigate the effect of timing of the reconstruction and radiotherapy, with respect to the complication rate and cosmetic outcome, with specific attention to the optimal timing of a definite implant in relation to the radiotherapy. 2. Methods The PUBMED database was searched for the period between January 2000 and December 2012. The following search strategy was used: (((((((((breast cancer)) OR (breast tumour)) OR (breast tumour)) OR (breast neoplasms)) OR (‘Breast Neoplasms’[Mesh]))) AND ((((radiotherapy)) OR (radiation therapy)) OR (‘Radiotherapy’[Mesh]))) AND ((((ablative surgery)) OR (mastectomy)) OR (‘Mastectomy’[Mesh]))) AND ((((breast reconstruction)) OR (‘Mammaplasty’[Mesh])) OR (mammaplasty)). Further, references from selected articles were searched as well. 2.1. Selection criteria of the studies

1) An immediate reconstruction has already been started by placing a tissue expander: A) Should the definite implant be placed before or after radiotherapy? 2) No reconstruction has yet been performed: A) Since implant reconstruction is not feasible if no tissue expanders have been placed, only an autologous reconstruction is possible. Should this be done before or after radiotherapy? 3) An autologous reconstruction has been performed: A) What is the effect of radiotherapy on the complication rate and cosmetic outcome? B) Is this effect different than if the reconstruction would have been performed after radiotherapy?

2.1.1. Design of studies and study population Articles had to be written in English or Dutch and were only included if they reported over at least 20 patients. Due to the lack of randomised trials, all study designs apart from case reports were included. We also included part of a study population if only that part fulfilled the selection criteria. We only included studies of patients with primary breast cancer. Studies concerning patients treated for recurrence or secondary breast cancer were excluded, since previous radiotherapy and/or surgery could bias the results. 2.1.2. Types of interventions Articles had to report the type of reconstruction and were divided into the category autologous or implant reconstruction. A reconstruction made with both

2754

J. Berbers et al. / European Journal of Cancer 50 (2014) 2752–2762

autologous material (Latissimus Dorsi Flap) and an implant, was categorised as a reconstruction with autologous material. They also had to report on the timing of the reconstruction: before or after radiotherapy. Since an implant reconstruction may either consist of two phases (first placing of a tissue expander (TE), followed by the definite implant), or one phase (immediate placement of the definite implant), we have categorised these different strategies. However, because placing either an implant or a TE after radiotherapy is almost impossible due to the radiotherapy-induced fibrosis, this strategy is hardly ever performed. Therefore, to maintain two separate categories, a definite implant placed after radiotherapy, with a tissue expander (TE) placed before radiotherapy, was categorised as reconstruction after radiotherapy. Only if the definite implant was placed before radiotherapy, it was categorised as reconstruction before radiotherapy.

of the study population, in relation to the total population for that study group. A second analysis was performed to minimise bias from interstudy variability using only those studies that directly compared reconstruction before with after radiotherapy, either for implant or for autologous reconstructions. This analysis was only made for complications that had two or more studies to compare. For the total complication rate, and the more serious complications like flap or implant failure, forest plots were made using Revman [22], comprising data from five of the seven studies that made direct comparisons. Patient and physician satisfaction on cosmetic outcome was also analysed per group using forest plots. Overall odds ratios and 95% confidence intervals were calculated and interpreted while taking into account the number and size of studies providing data.

2.1.3. Types of outcome measures The complication rates and reported rates of patient and physician satisfaction with cosmetic outcome were calculated. Capsular contracture and parenchymal fibrosis were combined as one category to be able to compare autologous reconstructions with implant reconstructions, since both complications are part of the same biological reaction on radiotherapy (the forming of fibrous tissue). When reporting on capsular contracture, we recorded the percentage capsular contracture mentioned in the article, or when graded according to the Spear–Baker classification (I–IV) [21], we scored grade III or IV as capsular contraction. We discarded acute radiation-induced complications such as desquamation and erythema, because we considered these side-effects to be independent of timing and type of surgery. Cosmetic outcome is represented in this review as patient and physician opinions about the cosmetic end result. We categorised their opinions as acceptable or unacceptable outcome. If an article reported cosmetic outcome as good, medium or bad, both good and medium were categorised as acceptable.

The Pubmed search yielded 641 hits. After applying the selection criteria only 37 articles remained (AvB, LB, JB). The selection concerned retrospective and prospective cohort studies; 28 papers concerned cohort studies of a specific subgroup of patients (e.g. all patients of the cohort were treated with an implant after radiotherapy, or all patients of the cohort had an autologous reconstruction before radiotherapy). In seven papers a direct comparison was made between reconstruction before or after radiotherapy, and in three papers a direct comparison was made between implant and autologous reconstruction. We did not find any randomised clinical trials on this subject. A total of 2683 patients were identified from these studies for metaanalysis, of whom 1635 were treated with autologous reconstruction, including 175 patients with an additional implant.

2.2. Statistics First, purely descriptive statistics was used to analyse the data. For each paper the type and incidence of complications were determined per complication and per study group: (1) Autologous reconstruction after radiotherapy; (2) Definite implant reconstruction after radiotherapy; (3) Autologous reconstruction before radiotherapy and (4) Definite implant reconstruction before radiotherapy. Subsequently, the weighted mean rates with their confidence intervals (CIs) were calculated assigning a weight to each individual study result based on the size

3. Results

3.1. Complications and satisfaction after radiotherapy and reconstruction in general Not all papers reported on the same type of complications. Most studies reported at least on infection, fibrosis, fat necrosis, and revision surgery; haematoma, seroma, and open wound were not always reported (Table 1). Most papers on autologous reconstruction gave some information on flap problems, whereas most papers concerning implant reconstructions reported on implant problems or failure. In 11 papers also patient and/or physician satisfaction with cosmetic outcome was reported (Table 1). Five out of 10 articles reported cosmetic outcome prospectively. The total average complication rate was 37%, with wide variation (8.7–70%). The most frequently occurring type of complications were fibrosis, implant problems, and ‘any complication requiring revision surgery’. Acceptable cosmetic

J. Berbers et al. / European Journal of Cancer 50 (2014) 2752–2762

2755

Table 1 Complication rates after the combination of mastectomy, reconstruction and radiotherapy in general, per study, and the weighted mean. Incidence in percentage* [ref] Haematoma Infection

Seroma Open wound Fibrosis (capsular contracture (implant)/ parenchymal fibrosis (autologous)) Fat necrosis

Flap problems (necrosis, partial necrosis) Flap failure Implant problems (e.g. implant deflation, leakage, dislocation) Implant failure (extrusion) Revision surgery

Complication rate (total)

Satisfaction of patient (acceptable) Satisfaction of physician(s) (acceptable) *

0 [33], 0 [34], 1.8 [35], 2.8 [28], 5.3 [36], 7.0 [37], 19.1 [38] 0 [34], 0.9 [39], 2.1 [29], 3.7 [11], 4.3 [28], 4.8 [40], 5.0 [31], 5.3 [36], 5.4 [13], 6.4 [38], 7.0 [37], 7.0 [41], 7.5 [35], 8.5 [42], 10.3 [33], 16.7 [43], 17.4 [44] 2.7 [13], 3.0 [41], 3.6 [35], 5.2 [28], 8.5 [42], 10.5 [36] 5.3 [36], 6.9 [45], 8.2 [33], 11.0 [41], 11.6 [29], 16.2 [13], 20.0 [46], 36.7 [43] 3.2 [40], 5.4 [28], 12.5 [35], 14.9 [38], 17.0 [47], 19.0 [48], 19.3 [12], 20.0 [49], 21.6 [13]22.7 [33], 23.5 [50], 32.5 [31], 32.6 [13], 36.0 [51], 39.0 [10], 39.7 [11], 55.0 [52], 56.7 [43], 57.8 [39] 1.0 [28], 3.6 [35], 8.0 [29], 8.5 [51], 9.0 [53], 12.0 [47], 12.0 [41], 15.1 [54], 15.4 [55], 16.2 [56], 19.6 [50], 23.3 [43], 23.7 [36], 25.0 [34], 47.0 [37] 0 [36], 1.0 [33], 3.0 [41], 4.1 [28], 6.9 [50], 7.0 [37], 16.0 [47], 37.5 [34] 0 [51], 0 [53], 0 [57], 0 [58], 1.0 [41], 2.1 [59], 2.2 [56], 3.0 [50], 7.0 [37], 10.1 [29] 0 [46], 2.5 [11], 3.1 [28], 4.3 [38], 8.3 [12] 0 [38], 1.6 [28], 2.5 [11], 8.3 [12], 10.0 [46], 15.9 [60], 16.2 [13], 19.6 [33], 22.7 [31], 25.2 [61], 17.0 [39], 44.2 [59] 0 [57], 0 [55], 0 [58], 2.9 [53], 6.4 [39], 7.5 [34], 9.0 [29], 9.0 [42], 10.0 [46], 11.1 [11], 17.6 [28], 27.7 [38], 35.8 [49], 45.5 [33], 47.0 [47], 54.0 [13], 66.7 [43] 8.7 [57], 9.0 [42], 11.1 [11], 20.0 [46], 26.0 [29], 27.9 [40], 28.5 [30], 30.8 [55], 32.1 [45], 37.1 [53], 37.5 [34], 37.7 [12], 40.0 [41], 45.9 [13], 48.1 [28], 48.9 [38], 50.0 [36], 58.8 [33], 70.0 [47] 41.4 [31], 67.0 [11], 67.5 [12], 80.0 [55], 82.6 [57], 83.0 [47], 88,4 [40], 89.7 [39], 93.3 [37] 55.8 [12], 62.2 [13], 80.0 [11], 82.8 [39], 83.0 [47]

Weighted mean INCIDENCE (%) 4 5

5 12 23

12

5 4 4 16 21

34

73 74

All incidence rates are presented in percentages.

outcome ranged from 41.4% to 93.3%. To investigate whether the variation in reported complications and acceptable cosmetic outcome could be ascribed to (1) timing of radiotherapy and (2) type of reconstruction, we analysed the studies using the subgroups mentioned above. 3.2. Outcome in relation to type and timing of reconstruction, with respect to the radiotherapy In our first analysis, based on comparing weighted means of the different groups, we found that the total complication rate was significantly higher for implant reconstructions if performed after radiotherapy (48.7%, 95%-CI 38.8–58.6%), than if performed before radiotherapy. (19.6%, CI 0.9–38.3%). The same holds true for revision surgery: 42.4% in case reconstruction was performed after radiotherapy (95%-CI 32.3– 52.5%) versus 8.5% if reconstruction was applied first (95%-CI 5.3–11.7%). However, for autologous reconstructions no statistical differences were found between total complication rates or revision surgery; only fibrosis occurred less frequently if this type of reconstruction was applied after radiotherapy (2.7%, 95%-CI 8.4– 13.0%) than if applied before radiotherapy (36.0%, 95%-CI 17.1–54.9%).

For the group with a reconstruction applied after radiotherapy, fibrosis occurred more frequently in the patients with an implant reconstruction (20.8%, 95%-CI 12.6–28.8%) than with an autologous reconstruction (2.7%, 95%-CI 0.5–9%). The same was seen for revision surgery: 42.4% with an implant reconstruction (95%-CI 32.3–52.5%) versus 11.5% with an autologous reconstruction (95%-CI 6.4–16.6%). In the reconstruction first group, autologous reconstruction seemed to yield a higher rate of revision surgery than the implant group (23.6% 95%-CI 5.7–41.5%, and 8.5% 95%-CI 5.3–11.7%). All other specific complications had overlapping results or insufficient data to make a fair comparison In our second analysis, where we confined the analysis to papers comparing total complication rate between reconstructions before or after radiotherapy (forest plots), no significant differences were seen, either for autologous reconstructions, or for implant reconstructions (Fig. 1). Concentrating on the more serious complications, i.e. flap failure or implant failure, no significant influence of timing of the reconstruction was seen on flap failure (Fig. 2). A significant higher rate of implant failures (OR 3.03 [1.59–5.77]) was seen if placed after radiotherapy, than if applied before radiotherapy (Fig. 3).

2756

J. Berbers et al. / European Journal of Cancer 50 (2014) 2752–2762

Fig. 1. Forest plot for total complication rates comparing reconstruction before or after radiotherapy, both for implant and autologous reconstruction. An odds ratio 1 favours reconstruction first.

Fig. 2. Forest plot for flap failure comparing reconstruction before or after radiotherapy, both for implant and autologous reconstruction. An odds ratio 1 favours reconstruction first.

Finally, we looked at patient and physician satisfaction on cosmetic outcome. Although an implant reconstruction after radiotherapy seemed to yield the lowest satisfaction rates, no large differences were seen between the four groups (Table 2). The forest plots showed no differences between the reconstruction before or after radiotherapy, for both patient satisfaction OR 1.12

[0.50–2.43] and physician [0.27–1.44] (Figs. 4 and 5).

satisfaction

OR

0.62

4. Discussion This review shows that there is an enormous variation in reported complication rates. Nevertheless, we

J. Berbers et al. / European Journal of Cancer 50 (2014) 2752–2762

2757

Fig. 3. Forest plot for implant failure comparing reconstruction before or after radiotherapy, both for implant and autologous reconstruction. An odds ratio 1 favours reconstruction first.

found an answer to the subject we investigated with special attention: the optimal timing of a definite implant reconstruction with respect to complication rate, is before radiotherapy, since total complication rate (weighted mean, not supported by forest plot), revision surgery (based on weighted mean, forest plot not available), and implant failure (forest plot, supported by weighted mean) occurred more frequently if an implant is applied after radiotherapy (Table 2 and Fig. 3). However, if the indication for radiotherapy is only known after a tissue-expander (TE) has been placed, delaying the radiotherapy until placement of the definite implant, may take that much time that it negatively influences local recurrence rate [23]. In that case, if adjuvant systemic treatment is indicated, this can be given during filling of the TE, followed by definite implant placement and radiotherapy. Nevertheless, an individual trade-off should be made between the potential risks on an increased recurrence rate by delaying radiotherapy, and risks of complications, by postponing the placement of the definite implant after radiotherapy. Further, we found no significant differences in total complication rate, or on serious complications like flap failure or revision surgery for autologous reconstruction performed before or after radiotherapy based on weighted mean or forest plots. Nevertheless, autologous reconstruction seems preferably to be performed after radiotherapy, since autologous reconstruction before radiotherapy results in more radiation-induced fibrosis (Table 2). Although this is expected to yield a worse cosmetic outcome, this did not appear in the forest plots in satisfaction with cosmetic outcome. However, the confidence intervals of all outcome parameters are wide.

Nevertheless, the fact that we found no hard data that the sequence of radiotherapy and autologous reconstructions has an impact on severe complications, other factors may play a role as well when choosing between radiotherapy first or reconstruction first. Immediate reconstruction for instance has technical and economic benefits: it is considered more practical to perform and is less costly than two separate operations [24]. Furthermore, direct reconstruction results in a better quality of life [7,8] since even a temporary loss of a breast can have significant psychosocial consequences. On the other hand, clinical experience suggests that an advantage of delayed reconstruction is that patients have time to consider what kind of reconstruction they want. The fear that mostly accompanies the diagnosis of breast cancer can interfere with the decision-making of reconstruction. Comparisons between the groups of implant and autologous reconstructions are even more difficult to make than comparisons between different timings of radiotherapy, because there are very limited data for each group and the differences between the groups are small. Other arguments to consider autologous versus implant reconstruction, are the habitus and preference of the patient; in some patients autologous reconstruction is not feasible [25,26]. A policy with shared-decision making is preferred, for which BRECONDA could be a helpful decision support tool [27]. Five other recent reviews concerning this subject have recently been published; the review of Barry and Kell [16] was limited to papers before 2010, whereas the others [17–20] reviewed the literature until May–October 2012, whereas our study extended to December 2012.

2758

Table 2 Complication rate and cosmetic outcome categorised by the timing of radiotherapy and subdivided in autologous and implant reconstructions.* Complications

Radiotherapy first

Reconstruction first

Autologous

Implant

Autologous

Implant

Weighted mean (95%-CI)**

Reported incidences

Weighted mean (95%-CI)**

Reported incidences

Weighted mean (95%-CI)**

Reported incidences

Weighted mean (95%-CI)**

Haematoma

0 [34], 3.2 [28]

3.0 (1.4–4.6)

4.5 (0–13.1)

/

/

0 [34], 2.1 [29], 3.7 [28], 7.0 [41]

3.6 (1.4–5.8)

6.8 (5.2–8.4)

0 [34], 5.3 [36], 7.0 [37] 0 [34], 5.3 [36], 7.0 [37], 8.5 [42], 16.7 [43]

3.6 (0–7.7)

Infection

8.1 (2.8–13.4)

0.9 [39], 3.7 [11], 30.0 [44]

3.5 (0–12.1)

Seroma

3.0 [41], 5.8 [28]

4.8 (2.2–7.4)

2.6 (0.7–4.5)

8.5 [42], 10.5 [36]

9.2 (7.4–11.0)

/

/

Open wound

7.0 [45], 11.0 [41], 11.6 [29] 0 [50], 3.7 [28]

10.7 (8.4–13.0)

10.3 (5.2–15.4)

5.3 [36], 36.7 [43]

23.1 (0–53.6)

20.0 [46]

20.0 (/)

20.8 (12.8–28.8)

17.0 [47], 23.5 [48], 36.0 [51], 56.7 [43], 75.0 [50]

36.0 (17.1–54.9)

0 [48], 19.3 [12], 39.0 [10], 39.7 [11], 55.0 [52], 57.8 [39]

39.8 (21.6–58.0)

Fat necrosis

1.1 [28], 8.0 [29], 8.6 [50], 12.0 [41], 13.3 [34], 16.0 [54], 16.2 [56]

8.9 (4.5–13.3)

0 [28], 0 [33], 1.8 [35], 19.1 [38] 4.8 [40], 5.0 [31], 5.4 [13], 6.4 [38], 7.5 [35], 7.7 [44], 9.1 [28], 10.3 [33] 0 [28], 2.7 [13], 3.6 [35] 0 [45], 8.2 [33], 16.2 [13] 3.2 [40], 12.5 [35], 13.6 [28], 14.9 [38], 21.6 [13], 22.7 [33], 32.5 [31] 0 [28], 3.5 [35]

2.5 (0–5.6)

18.7 (10.6–26.8)

/

/

Flap problems (necrosis, partial necrosis)

5.7 [50], 8.0 [28], 15.0 [41]

9.5 (4.6–14.4)

/

/

11.2 (5.9–16.5)

/

/

Flap failure

1.0 [41], 2.2 [56], 7.5 [59], 8.6 [50], 10.1 [29]

6.3 (2.5–10.1)

/

/

1.9 (0–4.5)

/

/

Implant problems (e.g. implant deflation, leakage, dislocation) Implant failure (extrusion)

/

/

4.3 [38], 7.1 [12], 27.3 [28]

10.9 (0–24.6)

8.5 [51], 9.0 [53], 12.0 [47], 13.0 [54], 15.4 [55], 23.3 [43], 23.7 [36], 32.0 [34], 43.8 [50], 47.0 [37] 0 [36], 7.0 [37], 9.4 [50], 12.0 [34], 16.0 [47] 0 [50], 0 [51], 0 [53], 0 [57], 0 [58], 7.0 [37], 7.5 [59] /

/

0 [46], 2.5 [11], 9.1 [12]

3.6 (0–7.5)

/

/

0 [38], 7.1 [12], 13.6 [28], 15.9 [60], 16.2 [13], 19.6 [33], 22.7 [31], 40.0 [39], 40.7 [59]

20.3 (13.1–27.5)

/

/

2.5 [11], 6.4 [39], 9.1 [12], 10.0 [46], 25.2 [61], 45.0 [59]

15.2 (5.6–25.8)

Fibrosis (capsular contracture (implant)/ parenchymal fibrosis (autologous))

2.7 (0–5.9)

J. Berbers et al. / European Journal of Cancer 50 (2014) 2752–2762

Reported incidences

83.7 (68.2–99.2) 42,9 [12], 80.0 [11], 92.1 [39] 72.1 (57.0–87.2)

78.2 (64.1–92.3) 42,9 [12], 67.0 [11], 80.0 [40], 88.9 [39] 81.7 (76.8–86.6)

11.1 [11], 20.0 [46], 48.0 [40], 62.5 [12] 36.8 (22.5–51.1)

19.6 (0.9–38.3)

6.4 [39], 10.0 [46], 11.1 [11]

73.5 (49.4–97.6) All incidence rates are presented in percentages. 95%-CI: 95% confidence interval. *

**

51.3 (/) 51.3 [12] Physician satisfaction

0 [12], 62.2 [13], 92.3 [39]

61.7 (33.8–89.6) 69.2 (/) 69.2 [12] Patient satisfaction

0 [12], 41.4 [31], 90.0 [40], 92.3 [39]

48.7 (38.8–58.6) 0 [12], 0 [45], 24.0 [40], 45.9 [13], 48.9 [38], 54.5 [28], 58.8 [33] 36.0 (28.2–43.8) 26.0 [29], 30.4 [12], 31.6 [45], 33.3 [34], 40.0 [41] 47.3 [28] Complication rate (total)

11.5 (6.4–16.6) 0 [34], 9.0 [29], 14.9 [28] Revision surgery

27.7 [38], 40.9 [28], 45.5 [33], 54.0 [13]

42.4 (32.3–52.5)

0 [55], 0 [57], 0 [58], 2.9 [53], 9.0 [42], 12.0 [34], 47.0 [47], 66.7 [43] 8.7 [57], 9.0 [42], 25.6 [30], 30.8 [55], 34.1 [12], 37.1 [53], 44.0 [34], 50.0 [36], 70.0 [47] 73.3 [12], 80.0 [55], 82.6 [57], 83.0 [47], 92.3 [37] 66.7 [12], 83.0 [47]

23.6 (5.7–41.5)

8.5 (5.3–11.7)

J. Berbers et al. / European Journal of Cancer 50 (2014) 2752–2762

2759

Although the reviews had a large overlap of reviewed papers, there were also quite a few differences. In addition, the angle and subsequent conclusions of these papers differed. The main conclusion of Barry and Kell [16] was that in case a reconstruction is performed before radiotherapy, autologous reconstruction is to be preferred above implant reconstruction. We could not confirm these data from our Table 2, but the confidence intervals are quite wide. Schaverian et al. [17] and Kelley et al. [18] focused on outcome after autologous reconstruction pre- or postradiotherapy. Schaverian et al. [17] support, and Kelly et al. [18] do not contradict our suggestion to preferably carry out an autologous reconstruction after radiotherapy. Schaverian et al. [17] came to this conclusion because less revision surgery was required than if performed prior to radiotherapy, which was also seen in Table 2, but did not reach significant difference. Kelly et al. [18] found no differences in outcome between pre- or post-reconstruction radiotherapy, but they did not report on fibrosis if radiotherapy is given prior to reconstruction, whereas fibrosis was our major reason to prefer radiotherapy first. Momoh et al. [19] and Lam et al. [20] focussed on the timing of radiotherapy in relation to implant reconstructions. However, the data of Momoh et al. [19] are not comparable, since they both considered radiotherapy on a tissue expander and on an implant as reconstruction first. The results of Lam et al. [20] corresponded with our data: they also saw more implant failures if the definite implant is placed after radiotherapy. The main limitation of this systematic review is that to our knowledge no randomised controlled trials have been published about this subject. Consequently, analyses of merely cohort studies can possibly lead to substantial bias with overrepresentation or underrepresentation of significant risk factors, such as obesity [28–30], peripheral vascular disease [29], hypertension [29], diabetes [29,30] and smoking [31]. This probably explains the wide variation in complication rates observed in different studies. Another explanation may be differences in personal/institutional specific experiences and differences in specific surgical techniques. Consequently, it may be worthwhile to analyse the data of one’s own clinic, when deciding on radiotherapy first or reconstruction first. In addition, most studies used in this review had relatively small sample sizes. Unfortunately it was not feasible to analyse studies of patients with a combination of an implant and autologous reconstruction separately, due to the small number of studies and patients in this group. Another problem we met in performing this meta-analysis, is that there is lack of standardised outcome complication parameters. Lastly, we did not analyse follow-up length. A short follow-up period may lead to underestimation of some complications and

2760

J. Berbers et al. / European Journal of Cancer 50 (2014) 2752–2762

Fig. 4. Forest plot for patient satisfaction comparing reconstruction before or after radiotherapy, both for implant and autologous reconstruction. An odds ratio 1 favours reconstruction first.

Fig. 5. Forest plot for physician satisfaction comparing reconstruction before or after radiotherapy, both for implant and autologous reconstruction. An odds ratio 1 favours reconstruction first.

overestimation of the cosmetic end result. Fibrosis only occurs after some time and cosmetic outcome can worsen over time. Two recent systematic reviews encountered similar problems regarding limited number and heterogeneity of published studies [16,32]. Further research in relation to timing of radiotherapy and type of reconstruction would be of great help to assess whether there is a difference between timing of radiotherapy and type of reconstruction in the incidence of complications and cosmetic outcome after the combination of mastectomy, reconstruction and radiotherapy. Large randomised controlled trials are preferable, using standardised outcome parameters. These RCTs are however probably hard to realise because of the still

relatively small number of patients receiving a combination of mastectomy, reconstruction, and radiotherapy. 5. Conclusion In general, timing of autologous reconstruction with respect to radiotherapy had no significant impact on total complication rate. Placing of a definite implant after radiotherapy seems to lead to a higher complication rate, with clearly more implant failures. Thus, if implant reconstruction is chosen, it is advised to be performed prior to radiotherapy. If autologous reconstruction is chosen, it seems slightly better to apply it after radiotherapy, since radiotherapy on an autologous

J. Berbers et al. / European Journal of Cancer 50 (2014) 2752–2762

flap may yield radiation-induced fibrosis with impaired cosmetic outcome. Further research and a more standard method of reporting of the results are necessary to provide more sustainable conclusions. Conflict of interest statement None declared. References [1] Jemal A et al. Global cancer statistics. CA Cancer J Clin 2011;61(2):69–90. [2] Clarke M et al. Effects of radiotherapy and of differences in the extent of surgery for early breast cancer on local recurrence and 15-year survival: an overview of the randomised trials. Lancet 2005;366(9503):2087. [3] Houssami N, Turner R, Morrow M. Preoperative magnetic resonance imaging in breast cancer: meta-analysis of surgical outcomes. Ann Surg 2013;257(2):249–55. [4] Painter TJ et al. Effect of magnetic resonance imaging on breast conservation therapy versus mastectomy: a review of the literature. Int J Surg Oncol 2011;2011. [5] Bride MBM et al. Factors associated with surgical decision making in women with early-stage breast cancer: a literature review. J Women’s Health 2013;22(3):236–42. [6] Smith KL, Isaacs C. BRCA mutation testing in determining breast cancer therapy. Cancer J 2011;17(6):492. [7] Al-Ghazal S et al. The psychological impact of immediate rather than delayed breast reconstruction. Eur J Surg Oncol 2000;26(1): 17–9. [8] Al-Ghazal S, Fallowfield L, Blamey R. Comparison of psychological aspects and patient satisfaction following breast conserving surgery, simple mastectomy and breast reconstruction. Eur J Cancer 2000;36(15):1938–43. [9] Recht A et al. Postmastectomy radiotherapy: clinical practice guidelines of the American society of clinical oncology. J Clin Oncol 2001;19(5):1539–69. [10] Contant C et al. Morbidity of immediate breast reconstruction (IBR) after mastectomy by a subpectorally placed silicone prosthesis: the adverse effect of radiotherapy. Eur J Surg Oncol 2000;26(4):344–50. [11] Cordeiro PG et al. Irradiation after immediate tissue expander/ implant breast reconstruction: outcomes, complications, aesthetic results, and satisfaction among 156 patients. Plast Reconstr Surg 2004;113(3):877. [12] Adesiyun TA et al. Impact of sequencing of postmastectomy radiotherapy and breast reconstruction on timing and rate of complications and patient satisfaction. Int J Radiat Oncol Biol Phys 2011;80(2):392–7. [13] Drucker-Zertuche M, Bargallo-Rocha E, Zamora-Del RR. Radiotherapy and immediate expander/implant breast reconstruction: should reconstruction be delayed? Breast J 2011;17(4): 365–70. [14] Kroll SS et al. Breast reconstruction with myocutaneous flaps in previously irradiated patients. Plast Reconstr Surg 1994;93(3): 460. [15] Evans G et al. Reconstruction and the radiated breast: is there a role for implants? Plast Reconstr Surg 1995;96(5):1111. [16] Barry M, Kell M. Radiotherapy and breast reconstruction: a meta-analysis. Breast Cancer Res Treat 2011;127(1):15–22. [17] Schaverien MV, Macmillan RD, McCulley SJ. Is immediate autologous breast reconstruction with postoperative radiotherapy good practice?: a systematic review of the literature. J Plast Reconstruct Aesthet Surg 2013;66(12):1637–51.

2761

[18] Kelley BP et al. A systematic review of morbidity associated with autologous breast reconstruction before and after exposure to radiotherapy: are current practices ideal? Ann Surg Oncol 2014:1–7. [19] Momoh AO et al. A systematic review of complications of implant-based breast reconstruction with prereconstruction and postreconstruction radiotherapy. Ann Surg Oncol 2014;21(1): 118–24. [20] Lam TC, Hsieh F, Boyages J. The effects of postmastectomy adjuvant radiotherapy on immediate two-stage prosthetic breast reconstruction: a systematic review. Plast Reconstr Surg 2013;132(3):511–8. [21] Spear SL, Baker Jr JL. Classification of capsular contracture after prosthetic breast reconstruction. Plast Reconstr Surg 1995;96(5): 1119–23. [22] Review Manager (RevMan) [Computer program]. Version 5.1. Copenhagen: The Nordic Cochrane Centre, T.C.C.; 2011. [23] Huang J et al. Does delay in starting treatment affect the outcomes of radiotherapy? A systematic review. J Clin Oncol 2003;21(3):555–63. [24] Veronesi P et al. Immediate breast reconstruction after mastectomy. Breast 2011;20:S104–7. [25] Singletary SE. Skin-sparing mastectomy with immediate breast reconstruction: the MD Anderson Cancer Center experience. Ann Surg Oncol 1996;3(4):411–6. [26] Malata C, McIntosh S, Purushotham A. Immediate breast reconstruction after mastectomy for cancer. Br J Surg 2000;87(11):1455–72. [27] Sherman K et al. BRECONDA: development and acceptability of an interactive decisional support tool for women considering breast reconstruction. Cancer Res 2012;72(24):S3. [28] Monrigal E et al. Mastectomy with immediate breast reconstruction after neoadjuvant chemotherapy and radiation therapy. A new option for patients with operable invasive breast cancer. Results of a 20 years single institution study. Eur J Surg Oncol 2011;37(10):864–70. [29] Baumann DP et al. Optimal timing of delayed free lower abdominal flap breast reconstruction after postmastectomy radiation therapy. Plast Reconstr Surg 2011;127(3):1100. [30] Berry T et al. Complication rates of radiation on tissue expander and autologous tissue breast reconstruction. Ann Surg Oncol 2010;17:202–10. [31] Cowen D et al. Immediate post-mastectomy breast reconstruction followed by radiotherapy: risk factors for complications. Breast Cancer Res Treat 2010;121(3):627–34. [32] Shah C, Kundu N, Arthur D. Radiation therapy following postmastectomy reconstruction: a systematic review. Ann Surg Oncol 2012:1–10. [33] Brooks S et al. Risk factors for complications of radiation therapy on tissue expander breast reconstructions. Breast J 2012;18(1):28–34. [34] Carlson GW et al. Effects of radiation therapy on pedicled transverse rectus abdominis myocutaneous flap breast reconstruction. Ann Plast Surg 2008;60(5):568. [35] Spear SL et al. Two-stage prosthetic breast reconstruction using alloderm including outcomes of different timings of radiotherapy. Plast Reconstr Surg 2012;130(1):1–9. [36] Spear SL et al. The effect of radiation on pedicled TRAM flap breast reconstruction: outcomes and implications. Plast Reconstr Surg 2005;115(1):84. [37] Halyard MY et al. Acute and chronic results of adjuvant radiotherapy after mastectomy and transverse rectus abdominis myocutaneous (TRAM) flap reconstruction for breast cancer. Am J Clin Oncol 2004;27(4):389. [38] Tallet AV et al. Radiotherapy and immediate two-stage breast reconstruction with a tissue expander and implant: complications and esthetic results. Int J Radiat Oncol Biol Phys 2003;57(1): 136–42.

2762

J. Berbers et al. / European Journal of Cancer 50 (2014) 2752–2762

[39] Nava MB et al. Outcome of different timings of radiotherapy in implant-based breast reconstructions. Plast Reconstr Surg 2011;128(2):353. [40] Anderson PR et al. Postmastectomy chest wall radiation to a temporary tissue expander or permanent breast implant – is there a difference in complication rates? Int J Radiat Oncol Biol Phys 2009;74(1):81–5. [41] Momoh AO et al. Delayed autologous breast reconstruction after postmastectomy radiation therapy: is there an optimal time? Ann Plast Surg 2012;69(1):14–8. [42] Wong JS et al. Incidence of major corrective surgery after postmastectomy breast reconstruction and radiation therapy. Breast J 2008;14(1):49–54. [43] Rogers NE, Allen RJ. Radiation effects on breast reconstruction with the deep inferior epigastric perforator flap. Plast Reconstr Surg 2002;109(6):1919. [44] Nahabedian MY et al. Infectious complications following breast reconstruction with expanders and implants. Plast Reconstr Surg 2003;112(2):467. [45] Kronowitz SJ et al. A multidisciplinary protocol for planned skinpreserving delayed breast reconstruction for patients with locally advanced breast cancer requiring postmastectomy radiation therapy: 3-year follow-up. Plast Reconstr Surg 2011;127(6):2154. [46] Salzberg CA et al. An 8-year experience of direct-to-implant immediate breast reconstruction using human acellular dermal matrix (AlloDerm). Plast Reconstr Surg 2011;127(2):514. [47] Albino FP et al. Irradiated autologous breast reconstructions: effects of patient factors and treatment variables. Plast Reconstr Surg 2010;126(1):12. [48] Whitfield GA et al. Incidence of severe capsular contracture following implant-based immediate breast reconstruction with or without postoperative chest wall radiotherapy using 40 Gray in 15 fractions. Radiother Oncol 2009;90(1):141–7. [49] Classen J et al. Fibrotic changes after postmastectomy radiotherapy and reconstructive surgery in breast cancer. Strahlenther Onkol 2010;186(11):630–6.

[50] Tran NV et al. Comparison of immediate and delayed free TRAM flap breast reconstruction in patients receiving postmastectomy radiation therapy. Plast Reconstr Surg 2001;108(1):78. [51] Huang CJ et al. Comparison of local recurrence and distant metastases between breast cancer patients after postmastectomy radiotherapy with and without immediate TRAM flap reconstruction. Plast Reconstr Surg 2006;118(5):1079. [52] Pinsolle V et al. Complications analysis of 266 immediate breast reconstructions. J Plast Reconstruct Aesthet Surg 2006;59(10): 1017–24. [53] Foster RD et al. Safety of immediate transverse rectus abdominis myocutaneous breast reconstruction for patients with locally advanced disease. Arch Surg 2005;140(2):196. [54] Peeters WJ et al. Fat necrosis in deep inferior epigastric perforator flaps: an ultrasound-based review of 202 cases. Plast Reconstr Surg 2009;124(6):1754. [55] McKeown D et al. The timing of autologous latissimus dorsi breast reconstruction and effect of radiotherapy on outcome. J Plast Reconstruct Aesthet Surg 2009;62(4):488–93. [56] Temple CLF et al. Choice of recipient vessels in delayed TRAM flap breast reconstruction after radiotherapy. Plast Reconstr Surg 2005;115(1):105. [57] Jhaveri JD et al. Clinical outcomes of postmastectomy radiation therapy after immediate breast reconstruction. Int J Radiat Oncol Biol Phys 2008;72(3):859–65. [58] Mehta VK, Goffinet D. Postmastectomy radiation therapy after TRAM flap breast reconstruction. Breast J 2004;10(2):118–22. [59] Nahabedian MY, Momen B. The impact of breast reconstruction on the oncologic efficacy of radiation therapy: a retrospective analysis. Ann Plast Surg 2008;60(3):244. [60] Seth AK et al. Utility of acellular dermis-assisted breast reconstruction in the setting of radiation: a comparative analysis. Plast Reconstr Surg 2012;130(4):750–8. [61] Ho A et al. Long-term outcomes in breast cancer patients undergoing immediate 2nd stage expander/implant reconstruction and postmastectomy radiation. Cancer 2011;118(9):2552–9.