Breast cancer affects one in eight women during their lifetime, and survival rates have been increasing.1 In 2020, approximately 275,000 women in the United States were diagnosed with invasive breast cancer, and approximately 48,500 were diagnosed with in situ disease.2 On average, 90 percent of these patients will survive for at least the next 5 years.2 Although mastectomy is among the most common forms of treatment for breast cancer, women often experience the long-term impact of this procedure on their body image and quality of life.
A higher proportion of women now elect to undergo breast reconstruction following mastectomy.3 A patient’s choice for their type of reconstruction (alloplastic, autologous, or a combination of both) is influenced by several factors, including patient preference, sociodemographic factors, and guidance provided by their reconstructive surgeon. In general, reconstructive surgeons can recreate a realistic breast mound and restore symmetry, often improving quality of life and patient satisfaction. As patient expectations have evolved over time, new techniques are being developed to not only restore breast appearance but also to reestablish breast sensation.4–7
During a mastectomy, sensory nerves (lateral and anterior branches of the second through sixth intercostal nerves)8,9 are transected, leaving diminished or complete loss of sensation in the remaining mastectomy skin flaps. Although reconstructed breasts and mastectomy skin flaps can spontaneously regain sensation, the recovery remains variable and unpredictable.5 Restoring breast sensation can have a positive effect on patient-reported quality of life,10 and can help a patient feel like their reconstructed breast is truly their own. The negative effects on psychological well-being from diminished or absent sensation following breast reconstruction were brought into the public eye in 2017 through the New York Times article “After Mastectomies, an Unexpected Blow: Numb New Breasts.”11
The impact of poor breast sensation following reconstruction has not gone unnoticed by the plastic surgery research community. One systematic review and four literature reviews have examined existing publications on breast sensation following implant-based and autologous reconstruction, with and without neurotization.5,12–15 Although these reviews reported that breast sensation after autologous reconstruction is likely superior,5,12 this conclusion is inconsistent with findings of subsequently published studies.16,17 A common challenge in breast sensation research is the difficulty to effectively interpret and synthesize published data because of different methods of evaluating breast sensation, and inconsistency in the timing of evaluation. These reviews acknowledge that traditional objective measures of sensation (e.g., Semmes-Weinstein/von Frey monofilaments, two-point discrimination) overlook the patient experience.4,5,12–15 Furthermore, the reviews draw attention to the need for a standardized approach to measure the experience and impact on quality of life of breast sensation from the patient’s perspective with meaningful patient-reported outcome measures.4,5,13,15
Although the existing BREAST-Q Physical Well-being of the Chest scale contains items that explore sensation of the reconstructed breast, it incompletely addresses sensation recovery and impact of sensation on quality of life.18 Thus, the BREAST-Q Sensation Module was developed and validated to comprehensively evaluate the impact of sensation on the reconstructed breast.19 The Sensation Module consists of three independently functioning scales: (1) Breast Symptoms, (2) Breast Sensation, and (3) Quality of Life Impact. As breast sensation after reconstruction should be a component of preoperative patient counseling,11,20 we evaluated whether sensation-related patient-reported outcomes differed between women who underwent alloplastic or autologous breast reconstruction postmastectomy.
PATIENTS AND METHODS
A sample of participants were recruited from the Love Research Army (formerly Army of Women) program, an online community created by the Dr. Susan Love Research Foundation to promote stakeholder participation in breast cancer research. Research ethics board approval for this study was obtained from Brigham and Women’s Hospital (Boston, Mass.) and the Love Research Army Scientific Advisory Committee (protocol number 2018P001073).
A recruitment e-mail was sent to Love Research Army members. Women were invited to participate if they were 18 years or older, had undergone mastectomy and breast reconstruction surgery, and were able to complete questionnaires in English. Women who met eligibility criteria were asked to provide information on sociodemographic and treatment characteristics and to complete the BREAST-Q Sensation Module scales through an online Research Electronic Data Capture21 survey.
To compare sensation-related patient-reported outcomes, data were analyzed only from patients who had either a “pure” autologous (i.e., abdominal, thigh, buttock, or latissimus flap without implant) reconstruction or a “pure” alloplastic breast reconstruction. Women who had mixed reconstruction (e.g., autologous reconstruction on one breast and alloplastic reconstruction on the contralateral breast) were excluded. Women who had undergone breast reconstruction less than 1 year previously were excluded, as the time needed for measurable sensory recovery of the reconstructed breast is estimated to be 12 months.22
BREAST-Q Sensation Module
The BREAST-Q, published in 2009,23 is a validated patient-reported outcome measure designed to measure quality of life (physical, psychosocial, and sexual well-being) and satisfaction (with breasts, outcome, and care) following different types of breast surgery. Participants completed the validated BREAST-Q Sensation Module,19 namely (1) Breast Symptoms (15 items), which asks about the experience (not at all, a little, quite a bit, a lot) of abnormal breast symptoms such as throbbing, burning, tingling, pressure, and others; (2) Breast Sensation (nine items), which assesses the amount of feeling in the breast (none, a little, some, a lot, complete) with items such as light touch, touch through clothing, erogenous touch, and others; and (3) Quality of Life Impact (eight items), which explores how much (not at all, a little, quite a bit, very much) loss of sensation in the breast has affected self-confidence, mood, intimacy, and others.19 Each scale results in an independent score from 0 (worst) to 100 (best). For each scale, a higher score means better quality of life, greater breast sensation, or fewer symptoms. The psychometric study reported high internal consistency (Cronbach’s alpha, 0.86 to 0.96) and test-retest reliability (intraclass correlation coefficient, 0.88 to 0.92).19
Descriptive statistics were used to describe cohorts. Patient, disease, and treatment characteristics were compared between alloplastic and autologous reconstruction groups using the chi-square test or Fisher exact test for categorical variables and the independent t test for continuous variables. Data were examined for violations of normality assumption before proceeding with parametric tests.
Multiple linear regression analysis was performed to identify factors that were independently associated (improved or worsened) with the three BREAST-Q sensation-related scale scores. Covariates in the models included patient variables (e.g., age, race, education) and treatment variables (e.g., time since last reconstruction, immediate or delayed reconstruction, radiation therapy). Background knowledge was used to guide variable selection to create the global model. Investigators used subject-specific knowledge and events per variable to derive the list of relevant variables. The role and properties of each of the variables were critically questioned. Statistical variable selection algorithms were not applied. Linear regression model assumptions pertaining to normality of the dependent variable and linearity were examined using a scatterplot. Normality and homoscedasticity of residuals were examined using a residual plot. Heteroscedasticity was addressed using robust standard errors in each regression model.
Regression adjustment was used to calculate the average treatment effect for each scale in the BREAST-Q Sensation Module (the average difference in scale score between alloplastic and autologous patients). Covariates in each regression model were selected based on statistical significance in the multiple linear regression analysis.
The Spearman rank-order coefficient was used to evaluate the strength of association between time since last reconstruction and each BREAST-Q Sensation Module scale’s score. The correlation coefficient was interpreted as follows: 0.00 to 0.30, negligible; 0.30 to 0.50, weak; 0.50 to 0.70, moderate; and greater than or equal to 0.70, strong association.24 We tested for differences in each BREAST-Q Sensation Module scale to explore the effect of time since the patient’s last reconstruction using the Jonckheere trend test. Statistical analyses were performed with STATA (StataCorp, College Station, Texas).25 The level of statistical significance was set at p < 0.05.
A total of 1204 women completed the survey. Of these respondents, 271 (22.5 percent) did not meet inclusion criteria for statistical analysis (Fig. 1). Demographic, clinical, and treatment characteristics of the 933 patients included for analysis are listed in Tables 1 and 2. The mean age of the participants was 59.2 ± 10.1 years, and the majority of women were White (93 percent), married (74 percent), and had attended or completed higher education (97 percent). Relative to the alloplastic reconstruction group, participants who underwent autologous reconstruction were older; had a higher body mass index and larger bra cup; and were more likely to have undergone unilateral mastectomy, radiotherapy, and immediate breast reconstruction (Tables 1 and 2).
Comparison of Patient Characteristics between Alloplastic and Autologous Reconstruction
|Characteristic||Total (%)||Alloplastic (%)||Autologous (%)||
|No. of patients||933||620 (66.5)||313 (33.6)|
|Mean age ± SD, yr||59.2 ± 10.1||58.1 ± 10.4||61.4 ± 9.1||<0.001|
|≤49 years||153 (16.4)||126 (20.3)||27 (8.6)|
|50–59 years||308 (33.0)||206 (33.2)||102 (32.6)|
|60–69 years||326 (34.9)||200 (32.3)||126 (40.3)|
|≥70 years||146 (15.7)||88 (14.2)||58 (18.5)|
|Mean ± SD, kg/m2||26.1 ± 5.2||25.5 ± 5.1||27.3 ± 5.2||<0.001|
|Normal (<25 kg/m2)||456 (48.9)||342 (55.2)||114 (36.4)|
|Overweight (25–30 kg/m2)||285 (30.6)||173 (27.9)||112 (35.8)|
|Obese (>30 kg/m2)||192 (20.6)||105 (16.9)||87 (27.8)|
|Mean age at first reconstruction ± SD, yr||48.9 ± 9.0||48.5 ± 9.2||49.7 ± 8.7||0.045|
|Time since last reconstruction||<0.001|
|1–2 years||145 (15.5)||97 (15.7)||48 (15.3)|
|3–5 years||175 (18.8)||134 (21.6)||41 (13.1)|
|6–10 years||335 (35.9)||245 (39.5)||90 (28.8)|
|11–15 years||137 (14.7)||83 (13.4)||54 (17.3)|
|≥16 years||141 (15.1)||61 (9.8)||80 (25.7)|
|A or smaller||59 (6.3)||35 (5.7)||24 (7.7)|
|B||296 (31.7)||211 (34.0)||85 (27.2)|
|C||360 (38.6)||248 (40)||112 (35.8)|
|D||159 (17.0)||98 (15.8)||61 (19.5)|
|DD and larger||59 (6.3)||28 (4.5)||31 (9.9)|
|White||870 (93.3)||588 (94.8)||282 (90.1)|
|Other||63 (6.8)||32 (5.2)||31 (9.9)|
|Married||694 (74.4)||457 (73.7)||237 (75.7)|
|Separated/divorced||122 (13.1)||80 (12.9)||42 (13.4)|
|Widowed||38 (4.1)||27 (4.4)||11 (3.5)|
|Single, never married||63 (6.8)||44 (7.1)||19 (6.1)|
|Other||16 (1.7)||12 (1.9)||4 (1.3)|
|Master’s/doctoral degree (complete and incomplete)||443 (47.5)||291 (46.9)||152 (48.6)|
|College, trade, or university diploma (complete and incomplete)||462 (49.5)||312 (50.3)||150 (47.9)|
|High school education or less||28 (3)||17 (2.7)||11 (3.5)|
Comparison of Disease and Treatment Characteristics between Alloplastic and Autologous Reconstruction
|Characteristic||All (%)||Alloplastic (%)||Autologous (%)||
|Invasive lobular and ductal carcinoma||288 (70.9)||154 (69.4)||134 (72.8)|
|Lobular and ductal carcinoma in situ||93 (22.9)||52 (23.4)||41 (22.3)|
|Other reason||25 (6.2)||16 (7.2)||9 (4.9)|
|Timing of reconstruction||<0.001|
|Immediate||528 (56.6)||317 (51.1)||211 (67.4)|
|Delayed||405 (43.4)||303 (48.9)||102 (32.6)|
|Unilateral||406 (43.5)||222 (35.8)||184 (58.8)|
|Bilateral||527 (56.5)||398 (64.2)||129 (41.2)|
|Yes||561 (60.1)||362 (58.4)||199 (63.6)|
|No||372 (39.9)||258 (41.6)||114 (36.4)|
|Yes||322 (34.5)||177 (28.6)||145 (46.3)|
|No||611 (65.5)||443 (71.5)||168 (53.7)|
|Yes||627 (67.2)||423 (68.2)||204 (65.2)|
|No||306 (32.8)||197 (31.8)||109 (34.8)|
|Yes||110 (11.8)||89 (14.4)||21 (6.7)|
|No||823 (88.2)||531 (85.6)||292 (93.3)|
|Tissue expander then implant||535 (86.7)||—|
|Patient did not recall/no response||4 (0.6)||—|
|Patient did not recall/no response||28 (4.5)||—|
|Abdominal flap (DIEP, free or pedicled TRAM)||—||282 (90.1)|
|Latissimus (without implants)||—||18 (5.8)|
|Gluteal (SGAP, IGAP)||—||4 (1.3)|
|Patient did not recall/other||—||4 (1.3)|
DIEP, deep inferior epigastric perforator; TRAM, transverse rectus abdominis musculocutaneous; SGAP, superior gluteal artery perforator; IGAP, inferior gluteal artery perforator.
BREAST-Q Sensation Module
Breast Symptoms Scale
The overall cohort reported a mean Breast Symptoms scale score of 78.7 ± 15.9. After adjusting for confounding variables, women who underwent autologous reconstruction scored an average of 6.1 points (95 percent CI, 3.9 to 8.4; p < 0.001) higher than alloplastic reconstruction respondents (Table 3).
Estimated Mean Difference in Score between Alloplastic and Autologous Reconstruction for BREAST-Q Sensation Module Scales*
|Scale||Unadjusted Mean ± SD||Adjusted Mean (95% CI)||
|Alloplastic (n = 619)||76.2 ± 15.6||76.5 (75.3–77.7)||<0.001|
|Autologous (n = 313)||83.5 ± 15.5||82.6 (80.7–84.6)||<0.001|
|Difference||7.3 (p < 0.001)
|Alloplastic (n = 612)||31.8 (19.4)||31.8 (30.3–33.4)||<0.001|
|Autologous (n = 305)||31.8 (22.8)||31.9 (29.3–34.4)||<0.001|
|Difference||0.0 (p = 0.98)
||−0.03 (−2.9 to 3.0)||0.754|
|Quality of Life Impact
|Alloplastic (n = 605)||63.8 (19.9)||64.0 (62.3–65.6)||<0.001|
|Autologous (n = 307)||70.6 (19.6)||69.3 (66.9–71.6)||<0.001|
|Difference||6.8 (p < 0.001)
BMI, body mass index.
†Included covariates: age, time since last reconstruction, radiation therapy, and education.
‡p value from the independent t test.
§Included covariates: marital status.
∥Included covariates: age, BMI, laterality, radiation therapy, hormonal therapy, and education.
Breast Sensation Scale
The overall cohort reported a mean score of 31.8 ± 20.5 for the Breast Sensation scale. After adjusting for confounding variables, no difference (0.0 points; 95 percent CI, −2.9 to 3.0; p = 0.75) was observed between autologous and alloplastic groups (Table 3).
Quality of Life Impact Scale
The overall cohort reported a mean score of 66.1 ± 20.4 for the Quality of Life Impact scale (Table 3). Women who underwent autologous reconstruction scored an average of 5.3 points (95 percent CI, 2.5 to 8.1; p < 0.001) higher than alloplastic reconstruction respondents.
Factors Independently Associated with BREAST-Q Sensation Module Outcomes
Table 4 displays results for the multiple linear regression analysis conducted for each scale to evaluate which factors influenced BREAST-Q Sensation Module scale scores. Respondent’s age was associated with improved Breast Symptoms and Quality of Life Impact scale scores by yearly increases of 0.2 (p = 0.003) and 0.35 (p < 0.001) points, respectively.
Patient and Treatment Factors That May Influence BREAST-Q Sensation Module Scale Scores*
|Breast Symptoms||Breast Sensation||Quality of Life Impact|
||0.1–0.3||–0.1||0.099||−0.3 to 0||0.3||0
|Time since last reconstruction (Ref., 1–2 years)|
|3–5 years||0.4||0.843||−3.2 to 3.9||0.9||0.689||−3.6 to 5.4||−1.0||0.641||−5.3 to 3.3|
|6–10 years||3.0||0.059||−0.1 to 6||1.1||0.604||−2.9 to 5.1||0.2||0.897||−3.5 to 4|
||1.5–8.9||3.1||0.221||−1.9 to 8||3.3||0.155||−1.3 to 7.9|
||2.3–9.9||5.8||0.051||0–11.7||0.9||0.731||−4.3 to 6.2|
|BMI (Ref., ≤25 kg/m2)|
|Overweight (25–30 kg/m2)||1.1||0.313||−1.1 to 3.4||2.1||0.198||−1.1 to 5.2||1.2||0.44||−1.8 to 4.1|
|Obese (≥30 kg/m2)||−2.1||0.103||−4.6 to 0.4||−0.5||0.795||−4.2 to 3.2||−4.5||0.011
||−7.9 to −1|
|Bra size (Ref., A)|
|B cup||−0.1||0.948||−4.5 to 4.2||0.7||0.838||−5.6 to 6.9||2.2||0.402||−3 to 7.4|
|C cup||1.2||0.581||−3 to 5.4||−0.8||0.796||−6.9 to 5.3||3.8||0.15||−1.4 to 9|
|D cup||−0.8||0.746||−5.4 to 3.9||−2.5||0.459||−9.1 to 4.1||0.4||0.878||−5.2 to 6.1|
|DD cup and larger||−0.7||0.802||−6.3 to 4.9||−3.8||0.354||−11.9 to 4.3||−1.6||0.661||−8.7 to 5.5|
|NSM (Ref., No)|
|Yes||−0.7||0.68||−4.2 to 2.7||−0.7||0.749||−4.9 to 3.5||3.8||0.08||−0.5 to 8.1|
|Surgery timing (Ref., Immediate)|
|Delayed||−0.4||0.69||−2.4 to 1.6||−0.5||0.742||−3.3 to 2.4||-0.4||0.789||−3 to 2.2|
|Laterality (Ref., Unilateral)|
|Bilateral||−2.0||0.066||−4.2 to 0.1||−1.2||0.454||−4.3 to 1.9||−4.3||0.002
||−7.1 to −1.6|
|Chemotherapy (Ref., No)|
|Yes||−0.5||0.672||−2.7 to 1.7||−2.1||0.179||−5.1 to 0.9||−2.2||0.133||−5 to 0.7|
|Radiation (Ref., No)|
||−5 to −0.6||0.4||0.825||−2.8 to 3.5||−3.1||0.03
||−6 to −0.3|
|Hormonal therapy (Ref., No)|
|Yes||−0.7||0.504||−2.9 to 1.4||−0.4||0.794||−3.4 to 2.6||−3.1||0.027
||−5.8 to −0.4|
|Marital status (Ref., Married)|
|Separated or divorced||0.3||0.866||−2.7 to 3.2||0.7||0.716||−3.2 to 4.6||−3.2||0.12||−7.1 to 0.8|
|Widowed||0.8||0.789||−5 to 6.5||3.6||0.344||−3.9 to 11.1||−3.6||0.367||−11.5 to 4.3|
|Single, never married||0.5||0.804||−3.4 to 4.4||7.2||0.013
||1.6–12.8||0.3||0.921||−4.9 to 5.4|
|Other||0.9||0.779||−5.5 to 7.3||-4.1||0.359||−12.8 to 4.6||0.7||0.902||−10.1 to 11.4|
|Education (Ref., College or university)|
|High school (complete and incomplete)||−2.0||0.527||−8.2 to 4.2||−3.0||0.517||−12.2 to 6.1||0.3||0.956||10.6–11.2|
|Masters/doctoral degree (complete and incomplete)||2.6||0.01
||0.6–4.6||1.2||0.377||−1.5 to 4||2.7||0.03
|Race (Ref., White)|
|Other||−1.8||0.363||−5.6 to 2||0.8||0.765||−4.7 to 6.3||−0.6||0.821||−5.5 to 4.3|
Coef., coefficient; Ref., reference; BMI, body mass index; NSM, nipple-sparing mastectomy.
*Multiple linear regression model including covariates listed above and reconstruction type were used to identify factors other than reconstruction type that significantly influenced BREAST-Q Sensation Module scale scores. Coefficient represents the average change in scale score for every one-unit change in covariate (continuous variables) or the average difference in scores compared to the reference value (categorical variables). The level of significance was set at p = 0.05. The value p = 0 represents a value of p < 0.001.
Breast Symptoms scale scores were higher (meaning few symptoms) in patients further away from their last reconstruction, with increases of 5.2 (p = 0.006) and 6.1 (p = 0.002) in patients 11 to 15 years and 16 to 20 years removed since their last reconstruction, respectively (Table 4). High body mass index and bilateral mastectomy had a negative effect on Quality of Life Impact scale, with scores decreasing by 4.5 (p = 0.011) and 4.3 (p = 0.002) respectively. Radiotherapy was associated with a modest decrease in Breast Symptoms and Quality of Life Impact scale scores. Across all three scales, bra cup size, nipple-sparing mastectomy, immediate reconstruction, and chemotherapy did not demonstrate a clear effect on sensation-related scale scores (Table 4).
Impact of Time Interval Since Last Reconstruction on BREAST-Q Sensation Module Scores
The median time since last reconstruction surgery was 8 years and was shorter in the alloplastic group than in the autologous group (p < 0.001). In both alloplastic and autologous reconstruction, most respondents [655 of 933 (70 percent)] had their last reconstruction within the past 10 years (Table 1). Breast Symptoms and Quality of Life Impact scale scores demonstrated a positive but negligible correlation with increasing time since last reconstruction in both alloplastic (rs = 0.21, p < 0.001 and rs = 0.12, p = 0.002, respectively) and autologous (rs = 0.22, p < 0.001 and rs = 0.19, p < 0.001, respectively) reconstruction (Fig. 2). (See Table, Supplemental Digital Content 1, which shows a summary of post hoc pairwise comparison using the Dunn test between BREAST-Q Sensation Module scale scores and time since breast reconstruction, https://links.lww.com/PRS/F448.) Breast Sensation scale scores had a positive but negligible correlation with time since last reconstruction for alloplastic patients (rs = 0.09; p = 0.0329). Breast Sensation scale scores did not appear to be affected by time since last reconstruction in autologous patients (rs = 0.017; p = 0.765).
This study reports the BREAST-Q Sensation Module outcomes for a large sample of women who underwent alloplastic or autologous breast reconstruction. We found that Breast Symptoms and Quality of Life Impact scale scores were higher (better) in patients undergoing autologous breast reconstruction (Table 3). Findings from this study also support that patient age and amount of time passed since last reconstruction are likely to have a positive effect on sensation-related outcomes, whereas exposure to radiotherapy confers a deleterious effect.
Although most of the literature investigating breast sensation focuses on autologous reconstruction,5,12–15 implant-based reconstruction is more common.26,27 This was also the case in our study, with 66 percent of our sample undergoing alloplastic reconstruction. Peer-reviewed literature directly comparing breast sensation between alloplastic and autologous reconstruction patients is rare,16 yet it is often implied that flap-based reconstruction provides better breast sensation than implants.5,12 Our analysis of 933 patients provides patient-reported data demonstrating Breast Symptoms scale scores (which reflect abnormal breast sensations, e.g. throbbing, burning) and Quality of Life Impact scale scores (which reflect, e.g., self-confidence, mood, intimacy) to be superior in autologous rather than alloplastic reconstruction.
In our study, the majority of alloplastic reconstruction patients had a tissue expander followed by an implant (86 percent), whereas the remainder were direct-to-implant (13 percent) reconstructions. After implant-based breast reconstruction, there is a significantly negative impact on the protective sensation of the breast skin compared to mastectomy without reconstruction.20 During mastectomy and reconstruction with tissue expander (or direct-to-implant), the nerve supply to the breast is disrupted. The exact mechanism behind sensory reinnervation of the reconstructed breast and overlying skin is unknown15; however, it is thought that the postmastectomy breast spontaneously regains sensation through ingrowth of sensory nerve branches from wound edges or from deeper structures.5 It is possible that ingrowth from “deeper structures” is prevented when alloplastic reconstruction is used.28 However, no difference in Breast Sensation scale scores were found (0.0; 95 percent CI, −2.9 to 3.0) between alloplastic and autologous reconstruction patients in our study.
Recently, immediate alloplastic breast reconstruction techniques involving nerve preservation and nipple-areola complex neurotization are being performed in an attempt to improve the return of breast sensation.7,29 The BREAST-Q Sensation Module provides a means to measure sensation-related outcomes from the patient’s perspective for these and future techniques. For clinicians and researchers incorporating objective breast sensation measurements into their clinical assessment, some reviews5,12 have suggested using the pressure-specified sensory device,30 a computer-assisted instrument, rather than Semmes-Weinstein monofilaments.
In our study, the majority of autologous reconstructions were abdominally based (92 percent). As responses were self-reported, the details of the surgical procedure (e.g., perforator versus muscle-sparing, neurotized versus nonneurotized flap) could not be obtained. Thus, the impact of the type of autologous flap on the recovery of breast sensation and the patient’s quality of life could not be ascertained. The first neurotized transverse rectus abdominis musculocutaneous flap was reported in 1992 by Slezak et al.31 Since then, one systematic review and three literature reviews agree that innervated flaps acquire superior sensory recovery, which starts at an earlier stage of recovery.5,12,13,15
Even after three decades of research and agreement of its superiority in the literature, flap neurotization in breast reconstruction has not been recognized or used as the standard of care.32 Many authors point to the inability to draw firm conclusions from published data because of a lack of standardized outcomes.5,14,15 Therefore, it is essential in this burgeoning frontier of microsurgical breast reconstruction33 to incorporate the validated BREAST-Q Sensation Module into future studies, making comparisons between studies, techniques, and flaps possible.
In the overall cohort, Breast Symptoms scale scores were better for patients who were further out from their last reconstruction. The abnormal breast sensations (e.g., throbbing, tingling, pulling) explored by this scale often improve over time; alternatively, patients may become desensitized to these symptoms or adjust to their new baseline. Many studies22,32,34 have demonstrated progressive sensory recovery over 1 to 2 years for autologous reconstruction patients. Although it is likely that return of breast sensation plateaus near the 2-year postoperative mark, this has not been demonstrated previously. Interestingly, our analysis comparing Breast Sensation scale scores between five postoperative time groups over a 20-year period demonstrated an increase, with a greater amount of time passed since last reconstruction for patients undergoing alloplastic reconstruction; however, the association was very weak and is unlikely to be clinically significant. Furthermore, when considering the implication of these results, it is important to note that this study was cross-sectional; thus, the trends in recovery of breast sensation at the patient level are unknown.
The independent factors included in the study that influenced the BREAST-Q sensation outcomes were consistent with the existing literature. Breast skin sensation diminishes with increasing age,35,36 and in our study, age was associated with improved Breast Symptoms scale scores. Radiotherapy is known to have a negative impact on the return of breast sensation and is a known predictor of persistent pain.37,38 This was further supported by our study.
We anticipated higher body mass index to have a negative influence on all three BREAST-Q Sensation Module scale scores. However, only the Quality of Life Impact scale demonstrated a meaningful and statistically significant effect.
The results of this study represent a sample of convenience, as members of the Love Research Army self-selected to participate. Furthermore, the study cohort is highly educated and White; therefore, it is unknown whether these results would apply to other populations. Clinical data were self-reported; consequently, accuracy of data could not be verified by cross-checking medical records. If patients had a neurotized flap, we assume they would be rare5 and thus the effect would be negligible in our sample size. Prepectoral reconstruction is a relatively new technique,39–41 and although it seems to have increased in popularity in the past few years, patients in our analysis had their reconstruction in 2018 or earlier, with the majority (85 percent) completing reconstruction before 2016.
As this is an observational study, statistical adjustments were made for known prognostic factors when analyzing data. However, as is common to all observational studies, there is potential for an imbalance of unknown prognostic factors between the alloplastic and autologous groups. In this study, there was also the inability to measure or control for certain variables that might influence sensation-related outcomes (e.g., the thickness of the mastectomy skin flaps). Furthermore, from an analysis standpoint, eliminating variables from the models based on statistical thresholds can potentially result in the omission of relevant confounders. (See Table, Supplemental Digital Content 2, which shows univariate analysis of patient and treatment factors that may influence BREAST-Q Sensation Module scale scores, https://links.lww.com/PRS/F449.) In addition, because we performed a cross-sectional analysis, we cannot establish a causal association between type of reconstruction and sensation-related outcomes.
Implications for Future Research
Existing reviews on breast sensation following reconstruction agree that the success of applying future research findings to benefit our patients will depend heavily on standardizing the way in which outcomes are assessed.5,12,13,15 Over the next decade, we anticipate a significant increase in the volume of research related to neurotized free flap breast reconstruction and alloplastic reconstruction using the prepectoral plane. Prepectoral implant placement decreases the amount of surgical dissection required, potentially disrupting fewer nerves.
Evaluating the sensory recovery of a reconstructed breast is complicated, as a normal breast consists of functional components (e.g., pain, temperature, vibration, sharp-dull) and erogenous sensation, and sensation can contribute to a patient’s experience of their reconstructed breast truly feeling like their own. Therefore, in addition to using objective breast sensation measurement devices to evaluate sensory recovery of the breast, we strongly recommend using the BREAST-Q Sensation Module to assess the impact of altered sensation from the patient’s point of view. Furthermore, breast sensation recovery is related to the amount of time passed since a patient’s last reconstructive procedure; consequently, we recommend analyzing sensation-related data a minimum of 1 year postoperatively.
Implications for Clinical Practice
This evaluation of reconstructed breast sensation using a validated patient-reported outcome measure provides generalizable data that can facilitate the breast reconstruction decision-making process. Educating and counseling patients preoperatively about the loss of normal breast sensation and the potential onset of new and unpleasant sensations following mastectomy is important. Our analysis suggests that reconstruction with autologous tissue may be associated with an improvement in abnormal breast sensations and quality of life. In the absence of surgical neurotization, there does not seem to be a difference in the return of breast sensation between autologous and alloplastic reconstruction. Although prospective research is required to confirm these findings, our results provide additional information to share with patients deciding between breast reconstruction modalities. Patients should be aware of all aspects of breast reconstruction, including breast sensation, before making their decision.