Fax: (416) 340-4403
Patient satisfaction and health-related quality of life after autologous tissue breast reconstruction
A prospective analysis of early postoperative outcomes
Article first published online: 24 OCT 2011
Copyright © 2011 American Cancer Society
Volume 118, Issue 6, pages 1701–1709, 15 March 2012
How to Cite
Zhong, T., McCarthy, C., Min, S., Zhang, J., Beber, B., Pusic, A. L. and Hofer, S. O. P. (2012), Patient satisfaction and health-related quality of life after autologous tissue breast reconstruction. Cancer, 118: 1701–1709. doi: 10.1002/cncr.26417
Fax: (416) 340-4403
- Issue published online: 2 MAR 2012
- Article first published online: 24 OCT 2011
- Manuscript Accepted: 9 JUN 2011
- Manuscript Revised: 6 JUN 2011
- Manuscript Received: 18 MAR 2011
- breast cancer;
- autologous tissue transfer;
For this study, the authors evaluated early psychosocial adjustments and health-related quality-of-life changes after breast reconstruction.
All consecutive patients who underwent breast reconstruction between June 2009 and November 2010 were asked to complete the BREAST-Q, Hospital Anxiety and Depression Scale (HADS), and Impact of Event Scale (IES) questionnaires before surgery and at 3 weeks and 3 months after surgery. A repeated-measures design was used to compare scores between baseline and postoperative time points.
Fifty-one of 55 women completed the questionnaires (response rate, 93%). BREAST-Q subscale scores (breast, sexual well being, and psychosocial well being) improved significantly (P < .05) postoperatively. The other subscale scores related to physical well being of the chest and abdomen dropped significantly 3 weeks after reconstruction; and, by 3 months after reconstruction, both scores improved significantly (P < .05). Large effect sizes for improvements in satisfaction, psychosocial well being, and sexual well being were observed (1.88, 1.2, and 1.31, respectively); whereas deterioration in the effect size for abdominal donor site was reported (−1.56). After adjusting for postoperative complications, there were statistically significant changes in BREAST-Q subscale scores. Changes observed on the HADS and IES provided external validation of the findings obtained on the BREAST-Q.
The current results suggested that the gains in breast satisfaction, psychosocial well being, and sexual well being after patients undergo either free muscle-sparing transverse rectus abdominis myocutaneous flap reconstruction or deep inferior epigastric artery flap reconstruction are statistically significant and clinically meaningful to the patient as early as 3 weeks after surgery. However, these gains are accompanied by significant deterioration in physical well being of the abdominal donor site. Cancer 2011;. © 2011 American Cancer Society.
Mastectomy remains a common surgical option for reasons related to both oncologic treatment and patient preference1-3 Because advances in adjuvant therapies have resulted in an increasing number of breast cancer survivors, both short-term and long-term consequences of surgical treatment on physical and psychosocial well being of breast cancer survivors are important. Although previous research has examined psychosocial outcomes among women who undergo mastectomy with and without reconstruction and breast-conservation therapy, many studies have used cross-sectional study designs, which lack baseline data for comparison and a longitudinal perspective.4-13
Recent studies that have sought to study health-related quality of life (Hr-QOL) issues after breast reconstruction using longitudinal study designs have generated mixed findings.14-17 Although most authors agree that postmastectomy breast reconstruction outcomes evolve with time,13, 15-23 there is no consensus regarding the psychosocial benefits in the early postreconstruction period.13-15, 24 The primary shortcoming of these studies is the inclusion of heterogeneous samples of patients who underwent a variety of different postmastectomy breast reconstruction techniques whose outcomes evolved with various timelines.
At our institution, the preferred reconstruction is a single-stage, microsurgically transferred, muscle-sparing (MS) transverse rectus abdominis myocutaneous (TRAM) flap or a deep inferior epigastric artery (DIEP) flap. Compared with the more invasive pedicled TRAM flap, which necessitates harvest and transfer of the entire rectus abdominus muscle, the MS-TRAM flap significantly limits the portion of muscle that is harvested. The DIEP flap is a further refinement of the MS-TRAM flap in which the overlying skin and subcutaneous tissues are solely perfused by 1 or more transmuscular branches. No muscle is harvested, which is hypothesized to further minimize potential donor site morbidity.
Because of the increasing popularity of the microsurgical, abdominally based reconstruction procedures combined with practical concerns about potential complications and lengthy recovery, the early psychosocial adjustments after these procedures are both important and timely. The objective of this study was to evaluate the impact of breast reconstruction using the MS-TRAM or DIEP flap on patient-reported satisfaction and Hr-QOL. By using the MS-TRAM or DIEP flap, we hypothesized that there would be an improvement in psychosocial well being and Hr-QOL in the early postreconstructive period compared with baseline measures.
MATERIALS AND METHODS
Research Ethics Board approval was granted for this study from the University Health Network, Toronto, Ontario, Canada. Patients were recruited from the University Health Network (Toronto General Hospital/Princess Margaret Hospital, Toronto, Ontario, Canada) from June 2009 through November 2010. Inclusion criteria were: women undergoing primary (immediate or delayed) free MS-TRAM or free DIEP flap procedures with 2 surgeons, age ≥18 years, and proficiency in English. Both unilateral and bilateral procedures were included. The choice of reconstructive option was based on the patient's previous radiation status, surgical incisions, body habitus, and both patient and surgeon preference.
After signed informed consent was obtained, patients completed 3 questionnaires. Data were collected preoperatively and at postoperative week 3 and month 3. All questionnaires were completed at home and were returned to the study coordinator by mail. Patient and treatment data were collected at baseline. Patient information included age, marital status, employment, income, education, and ethnicity. Treatment information included timing of surgery (immediate vs delayed), prior breast radiation treatment, and previous breast surgery. After surgery, surgical information was obtained from the electronic patient record on operative procedure, major perioperative complications, and need for adjuvant chemotherapy or radiotherapy.
The BREAST-Q Reconstruction Module is a patient-reported outcome measure that was developed to assess Hr-QOL and patient satisfaction after breast reconstruction.24-27 Four of the 6 subscales measure well being and satisfaction before and after reconstruction: 1) satisfaction with breasts, 2) psychosocial well being, 3) sexual well being, 4a) physical well being with respect to chest, and 4b) physical well being with respect to the abdomen donor site. Two additional subscales measure postreconstruction outcomes related to satisfaction with outcome and satisfaction with information. However, because we were assessing the changes in Hr-QOL and patient satisfaction after breast reconstruction, only the 4 subscales that included prereconstruction scores and postreconstruction scores were analyzed. All scales are scored from zero to 100, with higher scores indicating greater satisfaction or function. Good psychometric properties have been reported for the BREAST-Q subscales (Cronbach alpha from .88 to .96). Good test-retest reliability has been reported (intraclass correlation coefficient, 0.85-0.98).24-27
Hospital Anxiety and Depression Scale
The Hospital Anxiety and Depression Scale (HADS) measures symptoms of anxiety and depression.28 The HADS is a self-administered questionnaire consisting of 14 items; 7 items assess depressive symptoms, and 7 measure symptoms of anxiety. Scores range from zero to 21 for both anxiety and depression subscales, with higher scores indicating higher levels of symptoms. A score ≥11 on either subscale indicates that the responder is a “probable case for clinical anxiety or depression.” Internal consistencies reportedly are 0.84 for anxiety and 0.86 for depression.28
Impact of Event Scale
The Impact of Event Scale (IES) is a 15-item, cancer-specific distress measure.29 The IES assesses intrusive feelings and thoughts about cancer and the avoidance of these feelings and thoughts. The scale consists of 15 items (7 intrusion items and 8 avoidance items). A total IES score may be calculated (range, 0-75), or separate intrusion and avoidance subscale scores may calculated. Internal consistencies of 0.88 and 0.83 have been reported for intrusion and avoidance, respectively. Scores for the total IES can be divided into subclinical range (0-8), mild (9-25), moderate (26-43), and severe (≥44).29
Our selection criteria for timing and method of breast reconstruction were consistent with previously published guidelines.30-32 Immediate breast reconstruction usually was offered to women with a strong family history or gene positivity for breast cancer, in situ breast cancer, or clinical stage I or IIA breast cancer. The technique of breast reconstruction was selected based on location/type of cancer, medical and surgical risk factors, need for adjuvant radiotherapy, availability of local and distant donor tissue, desired size of the reconstructed breast, and patient preference. In general, autologous tissue reconstruction was offered to women who had a history of prior chest irradiation, available donor tissues, previous lumpectomies scars, and/or nonpliable chest wall soft tissues and to those who were medically fit. Our first choice for autologous tissue reconstruction was a free DIEP flap followed by a free MS-TRAM flap. The final decision for the type of reconstruction was made intraoperatively, depending on the caliber and intramuscular course of the dominant perforating vessel to the abdominal skin.33
Sample Size Calculation
The summary scores for the 2 BREAST-Q subscales (well being and satisfaction) are each transformed into a score between zero and 100. A clinically relevant change in the Hr-QOL has been defined as a difference that exceeds half a standard deviation of the baseline value.34 Because the standard deviation of both the psychosocial function and satisfaction distribution is approximately 20, the minimum significant difference in the score for each subscale is calculated as 10.24 When power is set at 85% with a β error of .15 and a standard α of .05, and the equation for a 1-sampled paired t test is applied using the minimum difference of 10, a minimum sample size of 36 patients was calculated for this study.
Descriptive data were calculated for the continuous variables (mean, standard deviation) and for categorical variables (frequency). The mean questionnaire scores were compared before and after breast reconstruction using 2-sided t tests for paired data (HADS and IES) or repeated-measures analysis (BREAST-Q). The repeated-measure Cohen d statistic was used to quantify the effect size of variables related to the BREAST-Q subscales. Cohen d is defined as the difference between 2 means divided by the pooled standard deviation for those means corrected for the correlation in which <0.2 indicates a “trivial” effect size, 0.2 to 0.5 indicates a “small” effect size, 0.5 to 0.8 indicates a “medium” effect size, and >0.8 indicates a “large” effect size.4 Because of the prospective nature of the current study in which preintervention scores (baseline) and postintervention scores were available for all patients, between-patient confounders like age, disease stage, or baseline scores did not require adjustment. However, we did perform an adjusted analysis for within-patient confounders that may have arisen during the study, such as a need for postoperative chemotherapy or postoperative complications. A subgroup analysis was performed between the immediate and delayed reconstruction groups to examine the differences in BREAST-Q scores at the 3 time points using independent sample t tests, to examine the main time effect using repeated-measures analysis, and to examine the interaction between time and procedure timing (immediate vs delayed). All analyses were performed using the SPSS statistical software package (version 19; SPSS Inc., Chicago, Ill) with 2-tailed tests of significance, and the significance level set at P < .05.
In total, 55 women completed the baseline questionnaires preoperatively. The mean patient age was 48 years (range, 28-77 years). Fifty-one women completed the questionnaire booklet at follow-up (93% response rate). Of the 51 women who completed the follow-up questionnaire, 46 (90%) underwent free DIEP flap reconstruction, and 5 (10%) underwent MS-TRAM flap reconstruction. Eleven patients (20%) had major postoperative complications after breast reconstruction. The descriptive data are presented in Table 1.
|Characteristic||No. of Patients (%)|
|Married or common-law||40 (78)|
|Highest level of education|
|≤High school||11 (22)|
|Graduate degree||5 (10)|
|Annual household income, $US|
|Employed full time or part time||36 (71)|
|Home maker/retired/others||14 (27)|
|South Asian or East Indian||4 (8)|
|Asian or Pacific Islander||2 (4)|
|White non-Hispanic||36 (71)|
|Long-term health condition|
|Laterality of reconstruction|
|Timing of reconstruction|
|Breast cancer diagnosis|
|In situ disease: LCIS/DCIS||15 (29)|
|Invasive breast cancer||32 (63)|
BREAST-Q subscale “satisfaction with breast,” “psychosocial well being,” and “sexual well being” scores were significantly higher at both postoperative times compared with baseline scores (P < .001) (Table 2). Both the chest and abdominal donor sites for the free MS-TRAM flap and the DIEP flap had significantly deteriorated scores at 3 weeks after surgery (P < .001). At 3 months after surgery, the chest physical well being score recovered to preoperative levels (P = .352). The abdominal physical well being score remained significantly lower at 3 months postoperatively compared with the preoperative score (P < .001) but was significantly higher than the score at 3 weeks postoperative (P < .001).
|BREAST-Q Score: Mean±SD|
|BREAST-Q Subscale||At Baseline, N = 55||3 Weeks After Reconstruction, N = 51||3 Months After Reconstruction, N = 51||P||Adjusted Pa|
|1. Satisfaction with breast||39.52±19.58||60.60±11.46||65.48±10.67||.000||.000|
|2. Psychosocial well being||54.27±18.14||68.31±16.64||75.04±15.49||.000||.002|
|3. Sexual well being||37.78±21.39||51.65±22.34||57.17±17.45||.001||.010|
|4a. Physical well being (chest)||74.26±12.50||62.41±12.53||76.70±12.78||.000||.000|
|4b. Physical well being (abdomen)||87.12±14.18||51.65±12.48||68.81±16.44||.000||.000|
The changes in BREAST-Q scores across the 4 subscales before and after reconstruction at the 2 postoperative time points are presented in Figure 1. To determine the clinical significance of the changes in patient-reported satisfaction and well being after reconstruction, effect sizes were calculated (Table 3).
|BREAST-Q Subscale||Mean Difference Between 3-Month and Baseline Scores||Cohen d Effect Size|
|Satisfaction with breast||25.96||1.88|
|Psychosocial well being||20.77||1.52|
|Sexual well being||19.39||1.31|
|Physical well being (chest)||2.44||0.26|
|Physical well being (abdomen)||−18.31||−1.56|
The improvement in satisfaction, psychosocial well being, and sexual well being at 3 months postoperatively all translated clinically to large effect sizes (1.88, 1.52, and 1.31, respectively). The increase in physical well being in the chest had a “small” effect size, and the deterioration in the abdominal donor site was indicative of a “large” effect size (Fig. 2).
An adjusted analysis to account for within-patient confounders was performed. None of our patients required chemotherapy after surgery. Eleven patients (20%) had postoperative complications after their reconstruction, which included hematoma (n = 5), flap venous congestion (n = 4), and mastectomy flap necrosis (n = 2). After adjusting for the presence of postoperative complications using a repeated-measures analysis, the change in scores across the BREAST-Q subscales remained statistically significant (Table 2).
Analyses on Timing of Reconstruction: Immediate Versus Delayed
At baseline, the delayed reconstruction group had significantly lower psychosocial functioning in all 3 BREAST-Q subscales compared with the immediate reconstruction group (Fig. 3) (satisfaction with breast subscale, P = .003; psychosocial well being subscale, P = .005; sexual well being subscale, P = .004). However, at the 2 postoperative times, there were no significant differences between the immediate and delayed groups on any of the 3 subscales. The repeated-measures analysis indicated that the main time effect was significant on the satisfaction with breast subscale (P < .001) and the psychosocial well being subscale (P = .005), and the main time effect approached significance on the sexual well being subscale (P = .053). In addition, there was a significant interaction over time according to the timing of the procedure on satisfaction with breast (P = .004), psychosocial well being (P = .041), and sexual well being (P = .001) (Fig. 3).
HADS and IES Scores
The HADS anxiety and depression subscale scores were all <11, indicating that these women did not report high levels of clinical anxiety or depression before or after surgery. The mean HADS anxiety subscale score decreased significantly at 3 weeks after reconstruction compared with the baseline score (P < .001), and there was no statistical difference in the mean HADS depression subscale score (P = .069) (Table 4). The mean level of postsurgical total cancer-related distress measured on the IES was significantly less at postoperative week 3 than the presurgical level (P = .009).
|Mean ± SD Score|
|Measure||At Baseline||3 Weeks After Reconstruction||P|
|Anxiety||7.90 ± 4.72||4.90 ± 3.36||.000*|
|Depression||3.02 ± 3.51||3.70 ± 3.20||.069|
|IES, n = 51|
|Total||26.45 ± 16.22||19.55 ± 16.50||.009*|
|Intrusion||14.12 ± 8.34||9.65 ± 8.06||.003*|
|Avoidance||12.33 ± 9.22||9.90 ± 9.55||.060|
Our current results indicate that gains in satisfaction, psychosocial well being, and sexual well being after MS-TRAM or DIEP flap reconstruction are statistically significant and clinically meaningful to the patient as early as 3 weeks after surgery. However, deterioration in physical well being related to the abdominal donor site is evident during the first 3 months after reconstruction. The goal of postmastectomy breast reconstruction is to restore the appearance of the breast and to improve psychological outcomes after cancer ablation.35-38 Thus, the assessment of patient outcomes using appropriately constructed and validated instruments is essential to evaluate and quantify the success of these surgeries from the patient's perspective.39, 40 By using specific patient-reported measures, such as the BREAST-Q, the HADS (for measuring general distress), and the IES (for capturing cancer-specific distress), we have analyzed Hr-QOL and satisfaction data in a prospective cohort of 55 patients who underwent free MS-TRAM or DIEP flap breast reconstruction.
We observed that, as early as 3 weeks after breast reconstruction, there were significant improvements (P < .001) in satisfaction with breast as well as psychosocial and sexual well being that continued to be observed at 3 months after reconstruction. Satisfaction with breast reconstruction and improvements in sexual and psychosocial well being were both clinically significant and meaningful to the patients, as indicated by the large effect sizes for the BREAST-Q subscale change scores. The HADS and IES scores provided external validation of these findings.
In our study, there were relatively low levels of anxiety and depression at baseline as assessed by the HADS and IES scores. A possible explanation why our patients were not overly anxious or depressed is that approximately 2 in 3 of our patients who underwent delayed breast reconstruction no longer had active breast cancer. In addition, HADS total scores (P < .001) as well as total IES scores (P = .009) were significantly lower as early as 3 weeks postoperatively compared with the scores before surgery. After adjusting for the presence of postoperative complications, we observed that the gains in satisfaction with breast, psychosocial well being, and sexual well being remained significant. This is consistent with findings reported by Cordeiro and McCarthy that patient satisfaction with breast reconstruction was high (95%) irrespective of postoperative complications.41 Our findings are in contrast to those reported by Andrade et al, who identified postoperative complication as an important indicator of patient dissatisfaction with reconstruction.42
The initial physical well being chest score declined significantly at 3 weeks postoperatively compared with the baseline score (P < .001), although the score returned to baseline levels by 3 months. The initial decline may be related to early postoperative discomfort associated with operative procedure. Similarly, there was a significant deterioration in the abdomen physical well being score at 3 weeks and at 3 months compared with the score before surgery (P < .001), and the effect size for the drop in score was large. A decrease in abdominal function during the postoperative recovery period was expected; however, the degree of abdominal dysfunction and dissatisfaction reported by these patients at 3 months after surgery was unexpected. In our series, 51 patients underwent free DIEP flap reconstruction, and only 4 patients underwent free MS-TRAM flap reconstruction. DIEP flaps were popularized because of their theoretical advantage of minimizing abdominal donor site morbidity by avoiding the harvest of the rectus abdominis muscle and anterior rectus fascia.43, 44 We observed that, despite our routine, careful intramuscular dissection during DIEP flap elevation, deterioration in self-reported abdominal physical well being was significant both statistically and clinically even at 3 months after surgery. Similar findings were reported in recent prospective studies in which a significant decrease in health status was observed (Medical Outcomes Study Short Form 36 physical health scores) in the early postoperative period after free DIEP or MS-TRAM flap reconstruction compared with preoperative scores.45, 46 This is in contrast to other studies, which reported nearly 100% recovery to complete activities of daily living after both free DIEP and free TRAM flap reconstruction.47-49 However, those studies did not include questionnaires that were developed using psychometric properties or validated in the breast cancer population. Data regarding the long-term recovery of the abdominal wall after free MS-TRAM or DIEP flaps from the patient's perspective were not evaluated in our study.
We observed that, at baseline, the patients who underwent delayed reconstruction compared with immediate reconstruction had significantly lower satisfaction with their breasts and psychosocial function. Breast reconstruction appeared to help the patients in the delayed reconstruction group reach the levels of satisfaction and psychosocial well being similar to the levels in the immediate reconstruction group after surgery. Although both groups improved after reconstruction, the delayed group improved by the third postoperative week, whereas improvement lagged behind in the immediate group at the second postoperative time point of 3 months.
We also observed a statistically significant and clinically meaningful increase in both satisfaction with breast and improvement in psychosocial and sexual well being in the early postoperative period after breast reconstruction. Overall, these results differed from the findings of several previous studies that reported an initial deterioration of psychosocial functioning and Hr-QOL after breast reconstruction.14, 15, 50 One major difference that may account for the differences in our results is that we restricted our analyses to patients who underwent only free tissue transfer using the free MS-TRAM or DIEP flap, whereas the other studies included all methods of reconstruction, and those methods may not be comparable to one another. It has been well documented that autologous tissue reconstruction using the abdomen generates higher patient satisfaction than exclusively implant-based or combined procedures.13, 19, 51, 52 In addition, among the autologous tissue reconstructions, abdominally based flaps had significantly higher general satisfaction (P = .011) and esthetic satisfaction (P = .016) than latissimus dorsi flaps.13 Finally, unlike implant-based or combined latissimus dorsi flap with implant reconstructions, which need first to temporize with a tissue expander followed by a second replacement surgery, the breast mound is completely created after free MS-TRAM or DIEP flap reconstruction. Therefore, it is conceivable that patient satisfaction and Hr-QOL scores would be lower in a patient who is still undergoing the process of breast mound creation than in someone who has completed it.
The strengths of this study include a prospective, repeated-measures design; a high response rate (93%); and the use of valid, reliable, procedure-specific, patient-reported outcome measures.4, 7, 8, 24, 53 The prospective study design provided longitudinal data and the opportunity to compare the progress and evolution of outcomes in our patient sample with each patient acting as her own control. In addition, because we designed this study to obtain measurements at specified time intervals after breast reconstruction (3 weeks and 3 months), this approach avoided the potential bias inherent in assessing Hr-QOL at various lengths of time after surgery.13, 15-23 Although the observed improvements in Hr-QOL domains and satisfaction in breast were both statistically and clinically significant after breast reconstruction, we did not use a control group. Without comparison groups who underwent mastectomy alone or breast-conservation therapy, the gains in Hr-QOL and satisfaction attributed to breast reconstruction cannot be fully placed into context. However, the results of this study demonstrate that reconstructive techniques like the free MS-TRAM or DIEP flaps, which are being performed with increasing frequency in many North American and European centers, produce significant improvements in psychosocial outcomes for the reconstructed patients compared with their presurgical levels. A future follow-up study examining Hr-QOL and psychosocial adjustments after breast reconstructive surgery will include breast-conservation therapy and mastectomy with and without reconstruction to increase the utility of the study findings.
Therefore, our outcomes reflect the experience of a single, high-volume, microsurgical breast reconstruction center and may not be generalizable to other settings. Although the objective of this study was to acquire a better understanding of the early temporal relation between breast reconstruction and Hr-QOL, longer follow-up time points are necessary to understand long-term outcomes, especially with regard to donor site morbidity.
In conclusion, to our knowledge, this is the first study to use the BREAST-Q, a newly developed, psychometric tested, procedure-specific, patient-reported outcome measure, in a prospective study before and after breast reconstruction. The results from this cohort suggest that the gains in satisfaction, psychosocial well being, and sexual well being after free MS-TRAM or DIEP flap reconstruction are significant and clinically meaningful to the patient as early as 3 weeks after reconstruction. However, there is a significant deterioration in physical well being of the abdominal donor site during the first 3 months after reconstruction. In light of the growing popularity of these microsurgical, abdominally based techniques and practical concerns about the extended recovery needed after these procedures, this information can provide breast cancer survivors with a more complete understanding of outcomes in the early postoperative period.
No specific funding was disclosed.
CONFLICT OF INTEREST DISCLOSURES
The authors made no disclosures.
- 26US Department of Health and Human Services, US Food and Drug Administration, Center for Drug Evaluation and Research, Center for Biologics Evaluation and Research, Center for Devices and Radiological Health. Guidance for Industry: Patient-Reported Outcome Measures: Use in Medical Product Development to Support Labeling Claims. Silver Spring, MD: US Food and Drug Administration; 2006.
- 27Satisfaction and quality of life in women who undergo breast surgery: a qualitative study [serial online]. BMC Womens Health. 2008; 9: 11., , , , .
- 30Optimal timing of breast reconstruction—an algorithm for management based on a review of the literature. Breast Dis. 2002; 16: 31-35., .
- 31Reconstruction of the breast. Surg Oncol Clin North Am. 1997; 6: 71-89., .
- 33Muscle sparing and free TRAM flaps. In: Hall-Findlay E, Evans G, editors. Aesthetic and Reconstructive Surgery of the Breast. London, United Kingdom: Elsevier Publishing Ltd.; 2010:93-115., , .