Medium-term cost analysis of breast reconstructions in a single Dutch centre: A comparison of implants, implants preceded by tissue expansion, LD transpositions and DIEP flaps

Background Free flap breast reconstruction (BR) is generally believed to be more expensive than implant BR, but costs were previously shown to level out over time due to complications and re-operations. The aim of this study was to assess the economic implications of four BR techniques: silicone prosthesis (SP), implant preceded by tissue expansion (TE/SP), latissimus dorsi transposition with or without implant (LD ± SP) and deep inferior epigastric perforator (DIEP) flap. Methods A prospective historic cohort study was performed to evaluate intramural medical costs in 427 patients, who had undergone BR between 2002 and 2009. Short- and medium-term complications were incorporated. In addition, 58 patients, who had recently undergone BR, participated in a questionnaire study to prospectively evaluate extramural medical and non-medical costs. Estimates of mean short- and medium-term costs are presented per patient. Results Intramural medical costs for BR and short-term complications for unilateral DIEP flaps (€12,848) and TE/SP reconstructions (€12,400) were significantly higher than those for LD ± SP reconstructions (€5804), which, in turn, were more expensive than SP reconstructions (€4731). In bilateral cases, costs of TE/SP (€12,723) and LD ± SP (€10,760) reconstructions were comparable, while DIEP flaps (€15,747) were significantly more expensive and SP reconstructions were significantly cheaper (€6784). Overall, the medium-term costs for complications and additional operations were not significantly different (€3017–€4503). Extramural medical costs and non-medical costs were approximately €9300 per stage, regardless of technique. Conclusions Differences in short-term costs between techniques did not level out during follow-up and SP reconstructions remained least expensive. Single-stage SP reconstructions, however, are not suitable for all patients due to high complication rates. Definite implant placement is therefore increasingly preceded by tissue expansion at more comparable costs to autologous BR. Incorporation of non-medical costs into the cost analysis would render two-stage procedures more costly than autologous BR. To achieve the optimal result, careful patient selection is critical. Only in select cases where two options are equally applicable, cost comparison becomes a valid argument for treatment selection. Keywords Breast reconstruction Complications Additional operations Cost analysis Societal perspective Different techniques are available for breast reconstruction (BR) after mastectomy. Commonly used techniques are silicone prostheses or implants (SP), implants preceded by tissue expansion (TE/SP), pedicled latissimus dorsi flaps with or without implant (LD ± SP), pedicled or free transverse rectus abdominis myocutaneous (TRAM) flaps or free deep inferior epigastric artery perforator (DIEP) flaps. Each technique can be performed directly after mastectomy (primary BR) or at a later stage (delayed or secondary BR). Multiple factors have to be considered to determine the optimal treatment modality for individual patients. Procedural characteristics have to be regarded in the context of patient-related factors (e.g., age, medical history and body habitus) and the surgeon’s expertise to assess which methods are technically feasible and which risks acceptable. Radiation therapy, for example, is a relative contraindication for implant BR due to increased complication rates, while sufficient excess abdominal tissue is a prerequisite for autologous BR using abdominal tissue. 1,2 When a multiplicity of techniques is available for the treatment of a particular disease, cost and outcome studies are important to demonstrate the procedures’ economic viability and ensure their unrestricted, continued availability. 3 In the Netherlands, for example, BR using a free flap is generally believed to be more expensive than implant BR, which provides implant-based BR a considerable advantage to insurance companies and might indirectly limit the availability of autologous BR. Previous studies have compared costs between implant and autologous BRs or between different types of autologous BRs. BRs using implants were (initially) less expensive than autologous BRs, 4 but costs levelled out within 5 years due to complications and re-operations. 5 These studies have contributed to valuable insight into costs and cost differences between different types of BR. However, some important health economic aspects were not incorporated properly: many studies used charges or fees rather than resource costs to estimate costs, 6–10 while a persistent relation between hospital charges and actual costs does not exist. 3,11,12 Furthermore, the hospital perspective was used, leaving extramural medical costs and non-medical costs unaccounted for. Finally, no studies have simultaneously assessed all available techniques. The aim of this study was to comprehensively assess the economic implications of four BR techniques (SP, TE/SP, LD ± SP and DIEP) from a societal perspective based on real resource use in substantial patient groups. Patients and methods Study design Mean total costs of SP, TE/SP, LD ± SP and DIEP flap BRs were studied from a societal perspective, meaning that medical as well as non-medical costs were considered. Medical costs include actual costs due to the medical treatment or its complications. A distinction was made between intramural medical costs (e.g., outpatient visits, inpatient days, surgery and blood tests) and extramural medical costs (e.g., general practitioner consultations, visits to physiotherapists and in-home nursing care). Non-medical costs included direct costs associated with patients’ travel costs and indirect costs resulting from absence from work (productivity costs). Intramural medical costs were evaluated in a prospective historical cohort study using patient charts and the electronic hospital administration. Extramural medical costs and direct and indirect non-medical costs were assessed using a questionnaire survey. The study was approved by the medical ethical review committee (MEC-2007-406). Intramural medical costs Patients, who had undergone primary or secondary BR at the Erasmus Medical Centre between 2002 and 2006, were included. The inclusion period for patients, who had received TE/SP reconstructions, was extended until March 2009, as this technique had only been performed sparsely between 2002 and 2006 and could otherwise not provide a representative study sample. Patients had been carefully selected to undergo the most appropriate type of BR. BR normally consists of multiple stages: breast mound creation, additional operations improving the aesthetic result or dealing with complications, nipple reconstruction and nipple areola complex (NAC) tattooing. 13 To evaluate patients who underwent comparable treatments, we distinguished between short- and medium-term costs. Short-term costs consisted of the initial breast mound reconstruction and short-term complications (<6 weeks). Costs for the mastectomy in case of primary BR and costs of both parts of two-stage procedures (TE/SP and bilateral LD ± SP) were included, as it was technically impossible to separate these costs. Medium-term costs consisted of medium- and long-term complications (>6 weeks) and additional operations to improve the aesthetic result, not including NAC tattooing. Mean total costs per patient were subsequently calculated by adding short- and medium-term costs. Short-term costs were analysed for all patients whose breast mound reconstruction had been completed. To be eligible for the medium-term part of the study, the patients’ entire BR process had to be completed, which was defined as a neo-breast with a reconstructed nipple or NAC. Patients without a nipple reconstruction were also included if they had sustained another additional operation but had refrained from further surgery 1 year after the last operation. The first 24 patients (30 flaps) of our DIEP flap series were excluded, as we showed previously that these patients represented our learning curve. 14 Extramural medical costs and direct and indirect non-medical costs Data regarding extramural health-care use, travel and sick leave were gathered prospectively using a standardised questionnaire, which was sent to patients, who underwent BR between November 2007 and August 2008. In addition to patients with SP, TE/SP, LD ± SP or DIEP flap BR, patients who underwent LD + SP preceded by tissue expansion (LD + TE/SP) were also included. In case of two-stage procedures, extramural medical costs were analysed for both stages separately. The questionnaire covered a 2-month period. Four weeks after mailing the questionnaire, non-responders were sent a reminder. One month later, the remaining non-responders were contacted by telephone. Patients who had undergone a two-stage procedure or respondents who had not yet fully returned to work were approached again after the second procedure or another 2-month period, respectively. Written informed consent was obtained from all patients who were contacted. Cost calculations Medical costs were calculated by multiplying volumes of health-care use with corresponding unit prices. For the most important intramural cost items, unit prices were determined by the hospital’s financial department, according to the micro-costing method. 15 Hospital salary schemes were used to estimate costs for personnel per hour. 16 Costs of equipment included those of investment, depreciation, interest and maintenance. It was deemed unnecessary to specify equipment use per surgical technique. Costs for equipment and most materials were calculated per hour of operative time. Only costs for tissue expanders and implants were specifically allocated to operations. Standard charges were used to calculate extramural medical costs and non-medical costs. 16 Productivity costs were only calculated for patients, who were employed at the time of BR. Only non-medical costs associated with the initial breast mound reconstruction and short-term complications were considered. Statistical analyses Analyses were performed per patient on an intention-to-treat basis: costs of non-assigned treatments and complications were accounted to the initial treatment group. Costs are presented in Euros and apply to the financial year 2006, when 1 € equalised approximately 1.27 US dollars. 17 If appropriate, unit prices were converted using the Consumer Price Index that corrects for inflation. Discounting was not relevant because of the limited time horizon (mean follow-up 5.4 years). To detect possible differences between different types of BRs, Pearson’s chi-square tests were used for categorical variables and Student’s t-tests or non-parametric Mann–Whitney U-tests for continuous variables. Subgroups were stratified by laterality (uni- or bilateral) and timing (primary or secondary) to allow accurate cost comparison between and within groups. Scatter plots and Spearman’s correlation coefficients were used to illustrate the relationship between medium-term costs and length of follow-up per patient. Two-tailed probabilities < 0.05 were accepted as statistically significant. Statistical analyses were performed using Statistical Package for Social Sciences (SPSS) version 15.0 (SPSS Inc., Chicago, IL, USA). Results Patient characteristics We included 427 patients with primary or secondary BR using one of four techniques for the analysis of intramural medical costs. No patients were excluded due to incomplete medical files. Patient characteristics are presented in Table 1 . Patients with SP or TE/SP reconstructions were significantly younger, underwent more bilateral and more primary BRs than patients with autologous reconstructions (p < 0.001). Short-term complications were more frequent in autologous reconstructions (p = 0.014), while medium-term complications, such as capsular contracture, implant displacement or abdominal herniation mostly involved implants (p < 0.001). In 35 patients, complications resulted in 42 failed reconstructions, of which 27 were salvaged using another type of BR. In addition, 76 patients who had recently undergone BR were contacted for the prospective questionnaire study to assess extramural medical costs and direct and indirect non-medical costs; the response rate was 76%. Non-responders were equally divided over the subgroups and did not differ significantly from responders with regard to age or complication rate. Sixteen patients were contacted again as they reported that they had not fully returned to work 2 months after the operation. Intramural medical costs Tables 2 and 3 provide a detailed overview of all cost components, unit prices and volumes of health-care use, which were used to calculate costs. Table 2 focusses on mean costs of the ‘standard’ initial operation, ‘without’ incorporating re-operations due to short-term complications. Mean costs per patient differed significantly between techniques (p < 0.001), with DIEP flaps being most costly (unilateral €5715; bilateral €8388), followed by TE/SP (€4832; €6966), LD ± SP (€2403; €5589) and SP (€2312; €3621) reconstructions. Costs for personnel (related to duration of surgery) and material (predominantly use of implant material) were accountable for most of these differences. Complications, re-operations and additional operations to improve the aesthetic result are incorporated in Table 3 , which presents short-term, medium-term as well as total costs. Short-term costs for unilateral DIEP flaps (€12,848) and TE/SP reconstructions (€12,400) did not differ (p = 0.792), and were significantly higher than those for LD ± SP reconstructions (€5804), which in turn were more expensive than SP reconstructions (€4731). In bilateral cases, costs of TE/SP (€12,723) and LD ± SP (€10,760) reconstructions were comparable (p = 0.136), while DIEP flaps (€15,747) were significantly more expensive and SP reconstructions were significantly cheaper (€6784). Costs for inpatient care and operations were the main causes for short-term cost differences. Medium-term costs after DIEP flaps (€4503; €3778), LD ± SP (€3757; €4179), SP (€3626; €3879) and TE/SP reconstructions (€3290; €3017) did not differ significantly (unilateral p = 0.551, bilateral p = 0.491). Costs of medium-term complications could not be separated from those of additional operations aimed at improving the aesthetic result. When analysing patients without medium-term complications separately, additional costs were significantly higher after DIEP flap BR than after other types of BR (unilateral p < 0.001, bilateral p = 0.013). A major part of this difference, however, can be attributed to one unilateral patient, who suffered complications after an additional procedure, resulting in total medium-term costs of €27,476. Exclusion of this patient, however, did not change results. Medium-term costs after SP reconstructions were significantly related to length of follow-up, with costs increasing over time (r = 0.230, p = 0.011; Figure 1 ). Extramural medical costs and non-medical costs Table 4 shows extramural medical and indirect medical and non-medical costs associated with the initial reconstruction. Costs related to absence from work (mean €9081) accounted for approximately 98% of extramural medical and non-medical costs (mean €9278). These costs did not differ between subgroups (p = 0.485). Discussion This study evaluated economic implications of four BR techniques. Short-term intramural medical costs differed significantly, with DIEP flaps being most costly per patient (unilateral €12,848; bilateral €15,747), closely followed by TE/SP (€12,400; €12,723), LD ± SP (€5804; €10,760) and SP reconstructions (€4731; €6784). Costs of unilateral DIEP flaps were comparable to unilateral TE/SP reconstructions, while bilateral LD transpositions, predominantly performed in two stages, were comparable to bilateral TE/SP reconstructions. Differences in short-term costs did not level out during follow-up (mean 5.4 years) and SP reconstructions remained the least expensive, costing on average €8357 for a completed unilateral reconstruction and €10,663 for a bilateral one. Single-stage SP reconstructions, however, are not suitable for all patients, and many authors would only apply this technique in distinct situations, depending on the size and shape of the (contralateral) breast as well as the quality of chest-wall soft tissues. 18 Radiation therapy and delayed reconstructions, amongst others, are considered contraindications. Internationally, there is a strong trend to expand overlying skin and muscle before placing the definite implant to improve the aesthetic result and reduce the risk for capsular contraction. 19–21 Only one of the authors (RTJW) has a vast personal experience with directly placing permanent implants, while the others (MAMM and SOPH) prefer to use tissue expanders. At our institution, implant BR is currently always preceded by tissue expansion. As expected, two-stage procedures resulted in higher costs than one-stage procedures. Costs of the standard initial breast mound reconstruction differed significantly between unilateral and bilateral TE/SP reconstructions (€4832 and €6966, respectively ( Table 2 )). This difference is obviously due to the use of a second tissue expander and silicone prosthesis and longer operating times. However, when complications are taken into account, mean total short-term costs are similar (unilateral €12,400; bilateral €12,723 ( Table 3 )). Out of 12 unilateral TE/SP reconstructions, three patients sustained short-term complications requiring one tertiary DIEP flap BR, compared to 8 out of 44 bilateral reconstructions requiring one tertiary LD + TE/SP reconstruction. Due to the relatively small number of unilateral cases, these short-term complications had a very clear impact on volumes of health-care use and, consequently, total short-term costs: operating times became comparable (unilateral 4.7 h; and bilateral 4.9 h) and hospital admittance was even longer (10.0 days; and 7.8 days). Furthermore, relatively more additional tissue expanders and silicone prostheses were used after unilateral than after bilateral reconstructions (7 and 11 implants, respectively). BR practice is subject to change. Current practice and, subsequently, its costs have already evolved from the situation presented here. The awareness raised by this study has already changed protocols. Patients are more liberally discharged with drains, which is primarily of financial benefit to autologous BRs. Furthermore, LD transpositions are increasingly combined with tissue expansion, which was not incorporated in this study, but will predictably result in higher costs. Timing of BR has an effect on overall costs. Even though our data did not allow accurate assessment of this effect, it was estimated by relating cost differences between primary and secondary reconstructions to costs of a separate mastectomy. For correct comparison, costs of mastectomy have to be added to those of secondary BR. In our study, primary reconstructions were generally more costly than secondary reconstructions (range €847–€2598; data not shown). In most cases, these differences did not exceed costs of mastectomy, which were estimated to be €2500, assuming that surgery takes 1 hour and patients are admitted for 3 days. These results support previously reported economic benefits of direct over delayed reconstructions, in addition to obvious personal benefits to the patient. 3,12,22 In our setting, primary BR is mostly performed in genetically predisposed patients undergoing prophylactic mastectomies; therapeutic mastectomies are mostly followed by secondary BRs due to both logistic difficulties as well as a cautious attitude by referring oncologic breast surgeons. Overall, medium-term costs were comparable between techniques and initial cost differences between autologous and implant reconstructions remained. DIEP flap patients underwent more costly additional operations than other patients. Patients who sustain autologous BR might be more demanding than those who undergo less invasive techniques and therefore pursue additional operations to improve the aesthetic result. We expected cost differences to (start to) level out, as the incidence of capsular contracture or implant displacement after implant BR is high and steadily increases over time. 23 Furthermore, symmetry problems due to changes in body weight and effects of gravity continue to develop later in life. Our follow-up period, however, was too short to incorporate all long-term complications. Longer follow-up is therefore required. Our medium-term results suggest an increase in additional costs after SP reconstructions over time, compared with more stable long-term costs after other types of BR. Longer recovery times and therefore higher productivity costs were anticipated after DIEP flap BR, while patients with implant BR were expected to return to work earlier. Surprisingly, patients returned to work approximately 10 weeks after surgery, regardless of type of BR. These recovery periods are comparable to those reported previously. 8 Our data suggest that the overall recovery period is longer after a two-stage procedure than after a one-stage procedure, even if the latter is more complex. Incorporation of extramural medical costs into long-term cost analysis would drastically change cost comparisons, especially as implant BR is increasingly preceded by TE. Sample size, however, might have been too small to detect differences. In addition, the Netherlands offers a very good social security system, which allows patients to take their time to return to work. Our data were obtained from a group of surgeons with extensive experience with BR within a single, large university hospital in the Netherlands. Our results are therefore not directly transferable to other health-care systems or less specialised surgeons. A teaching environment frequently mandates more medical staff to be involved than is common in general hospitals, rendering them more costly. By presenting our unit prices and volumes of health-care use in great detail, we enable others to translate our results to their own situation and adjust for experience, composition of surgical team or differences in technique. Costs of other types of BR, such as LD transpositions combined with tissue expansion, extended LD transpositions, superior-gluteal artery perforator (S-GAP) and transverse myocutaneous gracilis (TMG) flaps, can be estimated if the most important volumes of health-care use are known. Ideally, costs should be related to a measure of benefit. 5–7,12 The most important outcome to measure the effectiveness of BR would be patient satisfaction and quality of life. Patient satisfaction after BR is generally high, but several studies reported higher satisfaction rates after autologous reconstructions. 24–26 This study focussed on presenting costs in great detail. We are currently performing a prospective multicentre study assessing patient satisfaction and the psychological impact of different types of BR using multiple questionnaires. Incorporating such data in a cost analysis, rendering it a cost–utility analysis, would be the next step. Instead of patient satisfaction, the occurrence of complications can be used as an alternative outcome measure. Kroll et al. determined cost-effectiveness by correcting costs for the ultimate success rate, which was defined as the percentage of patients who achieved successful BR, even if that required conversion to another technique. 5,12 In our population, BR was ultimately unsuccessful in only 11 out of 427 patients (15 out of 597 reconstructions). An additional 23 patients required salvage (27 reconstructions) and another 102 patients required one or more re-operations due to complications. Unfortunately, such complicated cases are not incorporated in Kroll’s definition. Even though the number of complicated and salvaged reconstructions is significantly higher after implant BR, the number of failed reconstructions is comparable to other types of BR. Correcting costs for the ultimate success rate does therefore not change results drastically. Differences in methodology and local (financial) situation complicate direct comparison of our results to those of previous studies. In some studies, different types of BR (e.g., TE/SP and SP) were analysed together, 4,5,10 while others only presented standard treatment costs, disregarding costs caused by complications. 3,27 After converting all costs to 2006 Euros, our mean total costs for SP, TE/SP and DIEP flap reconstructions were comparable to those reported by Kroll. 5 In addition to directly comparing costs, volumes of health-care use can also be compared, thus bypassing differences in unit prices, charges or fees. Operative times were comparable to those reported in the literature, 1,4,5,22,27 but hospital admission after free flap surgery was (slightly) longer. 1,22,27 In our hospital, free flap patients are normally not discharged until all wound drains have been removed and patients have mobilised sufficiently. Especially in North America, discharge has been reported after 3–4 days. 1 Clearly, such differences in policy have a major effect on costs. Conclusion Short-term costs for unilateral DIEP flaps and TE/SP reconstructions were significantly higher than those for LD ± SP and SP reconstructions. In bilateral cases, costs of TE/SP and LD ± SP reconstructions were comparable, while DIEP flaps were significantly more and SP reconstructions significantly less expensive. Overall, medium-term costs were comparable between techniques and initial cost differences did not level out. However, longer follow-up is required. In current practice, definite implant placement is increasingly preceded by tissue expansion at more comparable costs to autologous BR. Subsequent incorporation of non-medical costs would further render two-stage procedures more costly than autologous BR. In addition to understanding the medium-term economic complications of BR from a societal perspective, professional assessment of the technical feasibility, acceptable risks and obtainable aesthetic result of different techniques will always remain of paramount importance to determine which technique is best suited for an individual patient. To achieve the optimal result, careful patient selection is critical, especially in case of single-stage SP reconstruction. Only in select cases where two options are equally applicable, cost comparison becomes a valid argument for treatment selection. Conflict of interest This study was not supported by any external sources or funds. The authors have no financial interest in any medical device, product or procedure mentioned in this article. Acknowledgements The authors would like to thank Paul Steinbusch and Niek Bossché from the Erasmus Medical Centre’s financial department for their help determining unit prices for most intramural cost items and Lenny van Moorsel-Spaans and Barbara Ponit from the Erasmus Medical Centre’s Business Intelligence Centre for their help in retrieving data from the hospital’s database. References 1 R.J. Allen P. Treece Deep inferior epigastric perforator flap for breast reconstruction Ann Plast Surg 32 1994 32 38 2 C.M. Contant A.N. van Geel B. van der Holt Morbidity of immediate breast reconstruction (IBR) after mastectomy by a subpectorally placed silicone prosthesis: the adverse effect of radiotherapy Eur J Surg Oncol 26 2000 344 350 3 M.J. Neyt P.N. Blondeel C.M. Morrison J.A. Albrecht Comparing the cost of delayed and immediate autologous breast reconstruction in Belgium Br J Plast Surg 58 2005 493 497 4 S.L. Spear S. Mardini J.C. Ganz Resource cost comparison of implant-based breast reconstruction versus TRAM flap breast reconstruction Plast Reconstr Surg 112 2003 101 105 5 S.S. Kroll G.R. Evans G.P. Reece Comparison of resource costs between implant-based and TRAM flap breast reconstruction Plast Reconstr Surg 97 1996 364 372 6 A. Thoma K. Veltri D. Khuthaila G. Rockwell E. Duku Comparison of the deep inferior epigastric perforator flap and free transverse rectus abdominis myocutaneous flap in postmastectomy reconstruction: a cost-effectiveness analysis Plast Reconstr Surg 113 2004 1650 1661 7 A. Thoma D. Khuthaila G. Rockwell K. Veltri Cost–utility analysis comparing free and pedicled TRAM flap for breast reconstruction Microsurgery 23 2003 287 295 8 J.M. Serletti S.L. Moran Free versus the pedicled TRAM flap: a cost comparison and outcome analysis Plast Reconstr Surg 100 1997 1418–24; discussion 1425–27 9 D.L. Larson N.J. Yousif R.K. Sinha J. Latoni T.G. Korkos A comparison of pedicled and free TRAM flaps for breast reconstruction in a single institution Plast Reconstr Surg 104 1999 674 680 10 A. Elkowitz S. Colen S. Slavin Various methods of breast reconstruction after mastectomy: an economic comparison Plast Reconstr Surg 92 1993 77 83 11 J.A. Cramer B. Spilker Quality of life and pharmacoeconomics: an introduction 1997 Lippincott Philadelphia 12 A. Khoo S.S. Kroll G.P. Reece A comparison of resource costs of immediate and delayed breast reconstruction Plast Reconstr Surg 101 1998 964–968; discussion 969–970 13 T.H.C. Damen M.A.M. Mureau H.A. Rakhorst S.O.P. Hofer The pleasing end result after DIEP flap breast reconstruction: a review of additional operations J Plast Reconstr Aesthet Surg 62 2009 71 76 14 S.O.P. Hofer T.H.C. Damen M.A.M. Mureau H.A. Rakhorst N.A. Roche A critical review of perioperative complications in 175 free deep inferior epigastric perforator flap breast reconstructions Ann Plastic Surg 59 2007 137 142 15 M. Gold J. Siegel L. Russel M.C. Weinstein Cost-effectiveness in health and medicine 1996 Oxford University Press New York 16 J.B. Oostenbrink C.A.M. Bouwmans M.A. Koopmanschap F.F.H. Rutten Handleiding voor kostenonderzoek, methoden en standaard kostprijzen voor economische evaluaties in de gezondheidszorg 2004 College voor Zorgverzekeringen, Geactualiseerde versie Diemen, The Netherlands 17 Available from: http://sdw.ecb.europa.eu/quickview.do?SERIES_KEY = 120.EXR.D.USD.EUR.SP00.A . 18 P.G. Cordeiro Breast reconstruction after surgery for breast cancer N Engl J Med 359 2008 1590 1601 19 P.G. Cordeiro C.M. McCarthy A single surgeon’s 12-year experience with tissue expander/implant breast reconstruction: part II. An analysis of long-term complications, aesthetic outcomes, and patient satisfaction Plast Reconstr Surg 118 2006 832 839 20 A.N. Mesbahi C.M. McCarthy J.J. Disa Breast reconstruction with prosthetic implants Cancer J 14 2008 230 235 21 L.L. Strock Two stage expander implant reconstruction: recent experience breast reconstruction supplement Plast Reconstr Surg 2009 22 M.H. Cheng J.Y. Lin B.G. Ulusal F.C. Wei Comparisons of resource costs and success rates between immediate and delayed breast reconstruction using DIEP or SIEA flaps under a well-controlled clinical trial Plast Reconstr Surg 117 2006 2139–2142; discussion 2143–2144 23 N. Handel T. Cordray J. Gutierrez J.A. Jensen A long-term study of outcomes, complications, and patient satisfaction with breast implants Plast Reconstr Surg 117 2006 757–767; discussion 768–772 24 A.K. Alderman E.G. Wilkins J.C. Lowery M. Kim J.A. Davis Determinants of patient satisfaction in postmastectomy breast reconstruction Plast Reconstr Surg 106 2000 769 776 25 Damen TH, Timman R, Kunst EH, et al. High satisfaction rates in women after DIEP flap breast reconstruction. J Plast Reconstr Aesthet Surg; 63:93–100. 26 Visser NJ, Damen TH, Timman R, Hofer SO, Mureau MA. Surgical results, aesthetic outcome, and patient satisfaction after microsurgical autologous breast reconstruction following failed implant reconstruction. Plast Reconstr Surg; 126:26–36. 27 S.S. Kroll G.P. Reece M.J. Miller Comparison of cost for DIEP and free TRAM flap breast reconstructions Plast Reconstr Surg 107 2001 1413–1416; discussion 1417–1418