Alderman, Amy K. M.D., M.P.H.; Bauer, Joseph M.D.; Fardo, Dean M.D.; Abrahamse, Paul Ph.D.; Pusic, Andrea M.D., M.S.
Breast augmentation is one of the most commonly performed elective procedures in the United States, and breast implants are one of the most frequently placed medical devices in all of medicine, with nearly 300,000 procedures performed annually.1,2 However, many health policy leaders have concerns about the safety of breast implants that have heightened with the European Poly Implant Prothèse implant crisis and the potential link of implants to anaplastic large cell lymphoma.3,4 High-quality, clinical outcomes research can help mitigate concerns about the effect of breast augmentation on women’s health, especially the safety and efficacy concerns expressed by the U.S. Food and Drug Administration.5 Furthermore, plastic surgeons need clinically meaningful, patient-centric outcomes data to optimize surgical outcomes.6
Traditional methods to evaluate surgical outcomes have included short- and long-term complications and aesthetic outcomes from the surgeon’s perspective. However, as plastic surgeons advance surgical techniques, the ways in which outcomes are assessed must also advance. Surgical outcomes need to be evaluated from the perspective of both the surgeon and the patient, realizing that the two perspectives may be very different. In addition, much of the previous research on patient satisfaction with breast augmentation has methodological limitations, such as the use of ad hoc questionnaires, which have limited reliability and validity.7–11 Fortunately, plastic surgeons now have access to a sensitive and specific patient-reported outcome instrument for elective and reconstructive breast surgery—the BREAST-Q—that can provide objective, reliable, and valid data.12–14 Use of the BREAST-Q may advance our understanding of breast surgery outcomes from the patient’s perspective that can be reproduced across different clinical settings and patient populations.
Our purpose was to prospectively evaluate patient-reported outcomes with elective breast augmentation using the previously validated BREAST-Q patient-reported outcome instrument. Specifically, we evaluated changes in patients’ satisfaction with their breasts and quality of life after elective breast augmentation. We also evaluated clinical factors that may be associated with satisfaction with breasts and with the overall surgical experience.
PATIENTS AND METHODS
We performed a prospective, single-center cohort study involving patients undergoing bilateral elective breast augmentation from January of 2011 to July of 2012. Before commencing our study, institutional review board approval was obtained. All consecutive female patients undergoing elective breast augmentation treated by one of three surgeons in our practice were offered study participation. All patients underwent breast augmentation using an inframammary incision and submuscular implant pocket. Patients with breast cancer or those with previous breast surgery were excluded.
Eligible patients were recruited for study participation at their preoperative visit and were asked to complete the self-administered breast augmentation module of the BREAST-Q survey. No incentive gifts were provided. A follow-up questionnaire about their surgical outcomes was administered to patients at the 6-week and 6-month postoperative visits. All patients agreed to participate in the study; however, 4 percent of patients were ineligible for study participation because of errors in survey administration (e.g., failure of the staff to administer the survey preoperatively or postoperatively). The number of patients who completed a preoperative questionnaire was 639; 90 percent (n = 576) of patients completed a postoperative survey at 6 weeks and 48.5 percent (n = 310) at 6 months.
The primary outcomes were patient satisfaction with breasts and quality of life measured with the BREAST-Q patient-reported outcome instrument.14 Development and validation of the BREAST-Q has been reported previously.12,14,15 In brief, the questionnaire was developed through a series of in-depth patient interviews, focus groups, expert panels, and literature review. Extensive field testing was performed for item reduction and final scale development across five centers in the United States and Canada. The BREAST-Q has a modular, procedure-specific structure for procedures involving breast reconstruction, augmentation, and mastopexy/reduction. The module for augmentation was used for this study. The scales of the BREAST-Q have proven reliability and validity. Reliability tests for internal consistency and test–retest reproducibility of the survey items. Internal consistency is measured using Cronbach’s α and item-total correlation statistical tests (>0.70 is considered acceptable for both tests). Test–retest reproducibility is evaluated using intraclass correlation coefficients and tests the stability of a scale with repeated administration (>0.70 is considered acceptable). The Cronbach’s α for the augmentation module ranges from 0.81 to 0.94, and the item-total correlations range from 0.55 to 0.82. Test–retest reliability, measured by intraclass correlation coefficients, ranges from 0.85 to 0.94.14
Satisfaction with breasts was assessed using a 17-item scale focusing on issues related to breast appearance, softness, amount of cleavage, size, scar appearance, and extent of implant rippling. Three quality-of-life subscales measured psychosocial well-being, sexual well-being, and physical well-being. The nine-item scale for psychosocial well-being addressed issues such as confidence in social settings, feeling attractive, and self-assurance. The five-item scale for sexual well-being addressed issues related to confidence with sexual activity and feelings of sexual attractiveness. The seven-item scale for physical functioning addressed issues related to pain in breasts, difficulty with lifting objects, and discomfort attributable to shifting of implants with physical activity. Satisfaction with the overall surgical experience was assessed using an eight-item scale addressing issues related to the overall experience, overall satisfaction, regrets, and extent that expectations were met. We also displayed some outcomes from the subset of questions that measured satisfaction with perioperative information, which was a 16-item scale that covered issues related to surgical risk, recovery, and outcomes. All questions were designed as a Likert scale response format. In accordance with the BREAST-Q Scoring Manual, patient scores were transformed into linearized, interval-level measurements using the Q-Score program and presented on a 0 to 100 scale, with higher values representing a more favorable outcome.16
The independent clinical variables obtained by chart review included patient age, height, weight, implant type (saline or silicone), implant volume, and surgeon. Age was analyzed as a four-level categorical variable (<30, 30 to 40, 40 to 50, and >50 years). Body mass index was calculated from the patient’s height and weight and reported as a three-level categorical variable (<25, 25 to 30, and >30). All implants were smooth round Mentor implants that were analyzed as either saline or gel-filled devices. The implant volume was calculated for each patient, taking an average of the implant volume from each side, and analyzed as a three-level categorical variable (<300, 300 to 400, and >400 cc). Varying the cutoff ranges for the categorical variables did not substantively change the results.
Descriptive statistics were performed to understand the demographic and clinical characteristics of our patient sample. In all analyses, preoperative body mass index was used rather than bra cup size because it appeared to be a more objective and reliable measure of patients’ preoperative body habitus size. t tests were used to compare preoperative and postoperative scores at 6 weeks and 6 months for satisfaction with breasts and quality-of-life subscales. To further assess outcomes related to physical functioning, an analysis of variance was performed to evaluate preoperative bivariate differences among the independent variables, and a logistic regression was used to assess physical functioning outcomes while controlling for patient and surgical factors. In the regression, a mean postoperative score of 70 or less indicated poor physical function (median score was 71). Varying the cutoff did not affect the results. We then calculated two standard indicators of effect size: (1) Kazis effect size17; and (2) standardized response mean.18 Larger effect sizes and standardized response means indicate greater responsiveness, and it is standard practice to interpret the magnitude using Cohen’s arbitrary criteria (0.2 for small, 0.5 for moderate, and 0.8 for large).19–21 We then calculated the mean score for satisfaction with breasts and satisfaction with the overall surgical experience across each independent variable. Logistic regression was used to evaluate the association of patient and surgical factors on satisfaction scores, controlling for the surgeon who performed the procedure. In the regression model, a score of 75 was considered “highly satisfied,” and varying the cutoff did not substantially change the results. Last, we described some specific information regarding individual questions within a scale. Questions within the satisfaction with the overall surgical experience used a three-point scale, and a t test was performed to show differences in responses by the independent patient or surgical variable. A p value less than 0.05 was considered statistically significant. All analyses were performed with STATA version 8.0 software (StataCorp, College Station, Texas).
Table 1 shows the study sample characteristics. The mean age was 33.5 years, and more than three-fourths of the patients were 40 years old or younger. The mean body mass index was 21.7, and 91 percent of patients had a body mass index less than 25. Nearly three-fourths of patients received a silicone gel–filled device, and the mean implant volume was 360 cc. More than two-thirds of patients had an implant fill volume between 300 and 400 cc.
Figure 1 displays changes in patient satisfaction with breasts and health-related quality of life with elective breast augmentation. Mean BREAST-Q scores ranged from 0 to 100, with higher scores indicating greater satisfaction or quality of life. Significant improvements are seen for all scales at both 6 weeks and 6 months postoperatively, except for physical functioning. Satisfaction with breasts improved from a mean preoperative score of 25.57 to 79.66 at 6 weeks (p < 0.001) and 84.10 at 6 months (p < 0.001), psychosocial well-being improved from a mean preoperative score of 51.86 to 85.40 at 6 weeks (p < 0.001) and 88.81 at 6 months (p < 0.001), and sexual well-being improved from a mean preoperative score of 38.45 to 82.54 at 6 weeks (p < 0.001) and 85.67 at 6 months (p < 0.001). However, physical functioning declined from a preoperative score of 94.83 to 71.04 at 6 weeks (p < 0.001) and 83.61 at 6 months (p < 0.001).
Table 2 shows the preoperative and postoperative 6-week physical functioning scores for the independent variables. In the bivariate analyses, advanced patient age, higher body mass index, and patients with saline-filled compared with silicone-filled breast implants had significantly worse preoperative physical functioning (p < 0.05). However, when preoperative physical functioning and patient-level factors were controlled for in the logistic regression model, postoperative physical functioning outcomes were not significantly correlated to the type of breast implant or fill volume.
Table 3 displays the magnitude of the effect of breast augmentation on patient satisfaction and health-related quality of life. Very large effect sizes are seen with all satisfaction and quality-of-life subscales. The Kazis effect size and standardized response mean are ways to measure how big of a difference a procedure has on a patient’s satisfaction or quality of life. Standard practice is to interpret a Kazis effect size or standardized response mean of 0.2 as small, 0.5 as moderate, and 0.8 as large. At both 6 weeks and 6 months postoperatively, the positive effect is greater than 2.5 for satisfaction with breasts (Kazis = 3.18 and 3.66, respectively; standardized response mean = 3.18 and 2.66, respectively), greater than 1.5 for psychosocial well-being (Kazis = 1.55 and 2.39, respectively; standardized response mean = 1.55 and 1.55, respectively), and greater than 2.0 for sexual well-being (Kazis = 2.37 and 2.56, respectively; standardized response mean = 2.37 and 2.23, respectively). Surgery appeared to have a negative effect on physical functioning, with an effect size greater than 2.0 at 6 weeks (Kazis = 2.30; standardized response mean = 2.30). However, this negative effect diminished at 6 months postoperatively (Kazis = 0.81; standardized response mean = 0.75).
Table 4 shows patient-reported satisfaction with breasts after augmentation mammaplasty, controlling for patient and surgical factors. These questions focused on breast size, feel, shape, and other aspects of physical appearance. Patient age was significantly associated with breast satisfaction and appears to diminish with age. Compared with patients younger than 30 years of age who had a mean satisfaction score of 82.0, the mean score for patients 30 to 40 years old was 80.1 (OR = 0.76, 95% CI = 0.50 to 1.16), for patients 40 to 50 years old was 78.2 (OR = 0.76, 95% CI = 0.44 to 1.31), and for patients older than 50 years was 71.1 (OR = 0.29, 95% CI = 0.12 to 0.77). Patient body mass index, implant type (saline versus silicone), and implant volume did not appear to be associated with postoperative satisfaction with breasts.
Table 5 shows patient satisfaction with the overall surgical experience after breast augmentation. These questions focused on the surgical experience, whether expectations were met, regrets about surgery, and satisfaction with the results. Patient body mass index and implant volume were not significantly associated with overall satisfaction with the surgical experience. However, patient age was associated with satisfaction with the overall experience and appeared to diminish with increased age. Compared with those younger than 30 years old who had a mean satisfaction score of 89.9, the mean score for patients 30 to 40 years old was 85.9 (OR = 0.69, 95% CI = 0.45 to 1.05), for patients 40 to 50 years old was 81.9 (OR = 0.60, 95% CI = 0.35 to 1.06), and for patients older than 50 years was 77.3 (OR = 0.24, 95% CI = 0.08 to 0.66). Patients with silicone implants also appeared to have diminished satisfaction, with a mean score of 85.1 compared with 88.3 for those with saline implants (OR = 0.62, 95% CI = 0.40 to 0.97). In particular, patients with silicone implants compared with those with saline implants were less likely to report that the surgery overall was a good experience (mean score of 2.88 versus 2.93, respectively, on a three-point scale; p < 0.05) and were less likely to feel that the surgery changed their lives for the better (mean score of 2.78 versus 2.89, respectively, on a three-point scale; p < 0.01). We explored reasons for the difference in satisfaction outcomes by implant type and did not find any differences in satisfaction with perioperative knowledge between patients with saline versus silicone implants regarding possible surgical complications, potential for implant leakage, risks associated with the implants, or the potential for rippling of the implant.
In this large, prospective cohort study of patients undergoing elective breast augmentation, we found significant improvements in women’s satisfaction with breasts, psychosocial well-being, and sexual well-being that persisted from the early postoperative period to 6 months afterward. Breast augmentation also has a very large magnitude of effect across all patient quality-of-life domains in both the early and later postoperative periods, except for physical functioning. Patients’ physical functioning does appear to be negatively affected by submuscular breast augmentation, although the negative effect is diminished at 6 months compared with 6 weeks. The physical functioning outcomes do not appear to be correlated to the type of breast implant (saline versus silicone) or fill volume. Furthermore, patient satisfaction with breasts postoperatively and with the overall surgical experience varies by patient age and implant type, with older patients less satisfied with both the surgical outcome and the surgical experience, and those with silicone gel–filled devices less satisfied with the overall surgical experience. However, there does not appear to be any differences in patients’ satisfaction regarding information about potential adverse events related to implant devices among those with saline versus silicone gel–filled devices.
Much of the quality-of-life research in breast surgery has focused on the improved quality-of-life changes associated with breast reconstruction or reduction mammaplasty and has been used to support health policy initiatives aimed at improving access to these procedures.22 Equally important is quality-of-life research with elective breast augmentation, especially given the heightened political concerns about the safety of these devices because of anaplastic large cell lymphoma awareness and the European Poly Implant Prothèse implant crisis.3,4 The U.S. Food and Drug Administration has stated that the efficacy of breast implants, not just the safety, must be documented in ongoing surveillance efforts.5 Validated measures such as the BREAST-Q can provide reliable and reproducible results addressing the efficacy and effectiveness of breast surgery. Our results, along with a recent report by Coriddi et al.,23 demonstrate that elective breast augmentation can significantly improve women’s quality of life. The psychosocial benefits include greatly improved confidence in social settings, feelings of attractiveness, and self-assurance. Confidence with sexual activity and feelings of sexual attractiveness are significantly improved. Patients undergoing the procedure are also significantly more satisfied with their breast appearance, size, softness, and amount of cleavage.
Our results also support those by McCarthy et al.,15 who found that elective breast augmentation has a significant and profoundly positive effect on women’s satisfaction with their breasts, along with their psychosocial and sexual well-being. Both our data and those of McCarthy et al. found very large (i.e., >0.80) effect sizes for each satisfaction and quality-of-life outcome. Other areas of surgery have evaluated the magnitude of effect a procedure can have on patients’ health-related quality of life. Hip arthroplasty has a documented large effect size of 3.1 on symptom improvement,18 pectus excavatum has a medium effect size of 0.6 on cardiopulmonary function,19 and carpal tunnel surgery has a small effect size of 0.2 on grip strength.24 When comparing across procedures, it is remarkable how large an effect size breast augmentation (Kazis effect size = 3.66) has on patient’s quality of life compared with those found for life-changing procedures, such as hip arthroplasty (Kazis effect size = 3.1) and pectus excavatum (Kazis effect size = 0.6). These results, along with those of McCarthy et al.,15 provide important data that can be used for both patient education and health policy initiatives regarding the positive effect of breast augmentation on women’s breast-related quality of life.
It is important to note the decline in physical functioning in our patient cohort. All patients in the study had a submuscular, not subglandular, augmentation. At 6 months postoperatively, physical function improved compared with 6 weeks postoperatively but did not return to baseline preoperative measures. A longer follow-up period is needed to assess whether physical functioning returns to preoperative levels or remains diminished in the long term, and a comparison cohort with a subglandular implant pocket would further elucidate the effect of pocket location on physical functioning. One could hypothesize that patients with a larger breast implant fill volume could have worse physical functioning compared with those with smaller implants, especially among more petite women with a lower body mass index. However, we found that postoperative physical functioning was not significantly correlated to the size of breast implant in either the bivariate or multivariate analyses. Research in the postmastectomy breast reconstruction population has not demonstrated any negative correlation between implant-based reconstruction and physical function outcomes,25 but data on physical functioning with elective breast augmentation are limited. These data are important for patient education and should be included when surgeons discuss with patients the implications of pocket location on mammography screening, aesthetic outcomes, and complications.
We found patient age to be significantly associated with satisfaction outcomes, with diminished satisfaction with advanced age. However, those older than 50 years still have a high satisfaction score of 71 of 100 for satisfaction with the postoperative breast result and 77 of 100 for satisfaction with the overall surgical experience. The older patients seem less satisfied with their aesthetic result and less likely to have their expectations met. We can only speculate as to the association between advanced age and diminished satisfaction. Older women have lived longer with micromastia and have learned to cope with issues related to body image, sexuality, and more pragmatic problems related to clothing fit. Older women may also have diminished body image related to the normal aging process, which may influence satisfaction with surgical outcomes. Similar findings have been found in the breast cancer population because older patients are less likely to receive postmastectomy reconstruction as a result of diminished desire for the procedures compared with younger patients.26 These data are important for surgeons to consider when counseling the older patient about elective breast augmentation but should not be viewed as a reason to deny this procedure because satisfaction outcomes are still very high.
The type of implant, saline versus silicone gel–filled devices, is an important decision for patients undergoing elective breast augmentation. Interestingly, we did not find implant type to be associated with satisfaction with the postoperative breast outcome. McCarthy et al.15 did find an association in the postmastectomy reconstructive population, with higher patient satisfaction with the postoperative breast result among patients with silicone compared with saline implants.27 We can only speculate, but the amount of soft-tissue coverage in the setting of elective breast augmentation compared with a postmastectomy defect may be enough to minimize differences in outcomes by implant type. It is important to note that the analyses controlled for patient body mass index, because thinner patients in our practice are more likely to receive a silicone gel–filled device. The reason why patients with silicone implants had diminished satisfaction with the overall surgical experience is unclear. We speculated that it was as a result of concern about implant safety; however, we found no differences in satisfaction with information about the devices among patients with saline and silicone implants. It is important to note that patients were not randomized to implant type. Thus, there may be other patient or clinical factors influencing the outcome that was not controlled for in the analyses. However, the differences in satisfaction among patients with saline versus silicone devices (mean score of 88.3 versus 85.1 of 100, respectively) were very small and may be more of statistical than clinical significance.
Our findings should be interpreted in the context of some limitations. Some patients were lost to follow-up at 6 months, and the outcomes of those patients are unknown. However, in our experience, highly satisfied patients are more likely to be lost to follow-up than dissatisfied patients, which would lead to an underreporting of improved satisfaction and quality-of-life outcomes. Furthermore, 6 months may be too short a follow-up period, especially for physical functioning outcomes. However, the other outcomes appeared relatively stable between 6 weeks and 6 months, suggesting that 6 months is adequate to assess early outcomes. Future BREAST-Q studies with follow-up beyond 6 months can be expected to provide additional important patient-centric outcomes data for patient education and regulatory efforts. Last, it may not be valid to generalize the results of this single group practice to other patient populations. Regional differences in patients and different surgical incisions, pocket locations, and device choices may lead to different patient-reported outcomes.
Our results have important implications for patient care and policy. Breast implants are one of the most frequently placed medical devices in all of medicine.1 Five to 10 million women worldwide have breast implants, and approximately 350,000 women in the United States undergo a breast implant procedure annually.2 Breast implants have been under close scrutiny by health policy leaders for decades, dating back to the silicone implant crisis of the 1990s,28 and have returned to the forefront of attention with the recent anaplastic large cell lymphoma concerns and the European Poly Implant Prothèse implant crisis.3,4 Some health policy leaders have taken a very negative view of these devices, claiming that they worsen the quality of life of women and therefore should be removed from the market.29–31 Our data suggest the opposite. Women with elective breast augmentation show significant improvements in psychosocial well-being, sexual well-being, and satisfaction with their breasts. Outcomes research on elective breast augmentation has never been more needed to ensure the continued availability of these devices in the marketplace. In particular, patient-reported outcomes provide a unique perspective of surgical outcomes that is becoming increasingly important in our consumer-driven healthcare system. It is no longer sufficient for physicians to report that patients are satisfied with surgery. The surgical experience must be documented from the perspective of the patient. These data along with other research using the BREAST-Q can be used on the policy level to support the availability of breast implants to women in the United States and worldwide.
The authors give special thanks to Collin Joiner for data management.
6. Chung KC, Swanson JA, Schmitz D, Sullivan D, Rohrich RJ. Introducing evidence-based medicine to plastic and reconstructive surgery. Plast Reconstr Surg. 2009;123:1385–1389
7. Park AJ, Chetty U, Watson AC. Patient satisfaction following insertion of silicone breast implants. Br J Plast Surg. 1996;49:515–518
8. Schlebusch L, Mahrt I. Long-term psychological sequelae of augmentation mammoplasty. S Afr Med J. 1993;83:267–271
9. Young VL, Nemecek JR, Nemecek DA. The efficacy of breast augmentation: Breast size increase, patient satisfaction, and psychological effects. Plast Reconstr Surg. 1994;94:958–969
10. Cash TF, Duel LA, Perkins LL. Women’s psychosocial outcomes of breast augmentation with silicone gel-filled implants: A 2-year prospective study. Plast Reconstr Surg. 2002;109:2112–2121 discussion 2122–2113
11. Swanson E. Prospective outcome study of 225 cases of breast augmentation. Plast Reconstr Surg. 2013;131:1158–1166
12. Pusic AL, Reavey PL, Klassen AF, Scott A, McCarthy C, Cano SJ. Measuring patient outcomes in breast augmentation: Introducing the BREAST-Q augmentation module. Clin Plast Surg. 2009;36:23–32, v
13. Klassen AF, Pusic AL, Scott A, Klok J, Cano SJ. Satisfaction and quality of life in women who undergo breast surgery: A qualitative study. BMC Womens Health. 2009;9:11
14. Pusic AL, Klassen AF, Scott AM, Klok JA, Cordeiro PG, Cano SJ. Development of a new patient-reported outcome measure for breast surgery: The BREAST-Q. Plast Reconstr Surg. 2009;124:345–353
15. McCarthy CM, Cano SJ, Klassen AF, et al. The magnitude of effect of cosmetic breast augmentation on patient satisfaction and health-related quality of life. Plast Reconstr Surg. 2012;130:218–223
17. Kazis LE, Anderson JJ, Meenan RF. Effect sizes for interpreting changes in health status. Med Care. 1989;27(3 Suppl):S178–S189
18. Liang MH, Fossel AH, Larson MG. Comparisons of five health status instruments for orthopedic evaluation. Med Care. 1990;28:632–642
19. Cohen J Statistical Power Analysis for the Behavioral Sciences. 19882nd ed Hillsdale, NJ Lawrence Erlbaum Associates
20. Cohen J. A power primer. Psychol Bull. 1992;112:155–159
21. Katz JN, Larson MG, Phillips CB, Fossel AH, Liang MH. Comparative measurement sensitivity of short and longer health status instruments. Med Care. 1992;30:917–925
22. Pusic AL, McCarthy C, Cano SJ, Klassen AF, Kerrigan CL. Clinical research in breast surgery: Reduction and postmastectomy reconstruction. Clin Plast Surg. 2008;35:215–226
23. Coriddi M, Angelos T, Nadeau M, Bennett M, Taylor A. Analysis of satisfaction and well-being in the short follow-up from breast augmentation using the BREAST-Q, a validated survey instrument. Aesthet Surg J. 2013;33:245–251
24. Cano SJ, O’Connor RJ, Thompson AJ, Hobart JC. Exploring disability rating scale responsiveness II: Do more response options help? Neurology. 2006;67:2056–2059
25. Alderman AK, Kuzon WM Jr, Wilkins EG. A two-year prospective analysis of trunk function in TRAM breast reconstructions. Plast Reconstr Surg. 2006;117:2131–2138
26. Alderman AK, Hawley ST, Morrow M, et al. Receipt of delayed breast reconstruction after mastectomy: Do women revisit the decision? Ann Surg Oncol. 2011;18:1748–1756
27. McCarthy CM, Klassen AF, Cano SJ, et al. Patient satisfaction with postmastectomy breast reconstruction: A comparison of saline and silicone implants. Cancer. 2010;116:5584–5591
28. . The breast implant crisis. In: History of the ASPRS. Plast Reconstr Surg. 1994;94:91A–99A