Abstract
Background:
The association between textured surface breast implants and breast implant-associated anaplastic large cell lymphoma (BIA-ALCL) has led to an increase in surgical procedures to exchange textured devices to smooth surface implants. Because patient satisfaction is an integral part of breast reconstruction, the purpose of this study was to compare patient-reported outcomes between smooth and textured implant recipients.
Methods:
Patients 18 years of age or older who underwent implant-based postmastectomy breast reconstruction with either smooth or textured devices from 2009 to 2017 and completed the BREAST-Q patient-reported outcome measure following reconstruction were included in this analysis. The primary outcomes of interest were mean and median BREAST-Q scores and postoperative complications.
Results:
Overall, 1,077 patients were included—785 underwent breast reconstruction with smooth implants and 292 with textured implants. No statistical differences were observed between the textured and smooth implant groups for any of the BREAST-Q domain scores during the early (3-month) to late (2-year) postoperative time points. Smooth implant recipients reported significantly more rippling (p=0.003) than textured implant recipients. In contrast, textured implant recipients had a higher rate of cellulitis than smooth implant recipients (p=0.016).
Conclusions:
These data suggest that postoperative satisfaction with breasts or health-related quality of life following immediate postmastectomy implant-based breast reconstruction is likely independent of implant surface type. However, smooth breast implants may result in more rippling. Our findings represent an important aid in counseling patients who have questions about the risks and benefits of replacing their textured implants with smooth surface devices.
INTRODUCTION
Implants are the most common method used for breast reconstruction in the United States.(1) The discovery that implants with textured surfaces are associated with the development of breast implant-associated anaplastic large cell lymphoma (BIA-ALCL) has led to considerable debate about the continued use of these devices.(2) While some advocate that the use of textured implants is justified by perceived improved cosmetic and overall outcomes, others suggest that textured implants should no longer be used due to the risk of BIA-ALCL.(3-5) Although the FDA has not recommended explantation of textured implants in asymptomatic patients,(6, 7) many patients are concerned about the risk of developing BIA-ALCL and request guidance regarding the risks and benefits of exchanging their implants from textured to smooth.(7, 8)
Advocates of textured breast implants argue that these devices decrease the rate of device malposition, reduce the rate of capsular contracture, and have superior cosmetic outcomes than smooth breast implants.(9-14) Textured breast implants, unlike smooth devices, are also available in anatomic shapes that are thought to result in more natural reconstructive outcomes.(15, 16) However, the scientific evidence for some of these claims is scant. Although some early studies showed decreased rates of capsular contracture with textured breast implants, others show similar rates when comparing textured and smooth breast implants.(17-21) In a small study of 120 patients, MacAdam et al. found no difference in patient-reported outcomes between women who underwent breast reconstruction with textured/anatomic devices with those who underwent reconstruction with smooth round implants.(22) Other studies also found similar BREAST-Q scores in patients with round or anatomic breast implants, but they likewise do not exclusively compare smooth and textured implants.(23-25)
The purpose of this study was to compare both patient-reported outcomes and complication profiles in a large cohort of patients who underwent postmastectomy breast reconstruction using textured or smooth breast implants. We hypothesize that there is no difference in outcomes between textured and smooth devices.
METHODS
Data Collection and Patient Variables
After approval from the institutional review board of Memorial Sloan Kettering Cancer Center (MSK), we reviewed electronic medical records from January 2009 to July 2017 for Women 18 years of age or older who underwent immediate breast reconstruction with tissue expanders and subsequent exchange to smooth or textured silicone breast implants following prophylactic or therapeutic mastectomy.
Data were retrospectively collected from prospectively maintained patients' electronic records of postoperative clinic and urgent care visits. Patients were stratified into two groups by implant surface type: smooth and textured. Demographic data, treatment method, and postoperative outcomes were recorded and compared between the smooth and textured implant groups. Variables recorded for each patient included age, race/ethnicity, smoking history, body mass index (BMI), diabetes, hypertension, breast malignancy history, radiotherapy status (none, pre- and post-tissue expander placement), chemotherapy status, timing of reconstruction, and laterality of reconstruction. Patients were noted to have a complication if the complication was documented postoperatively in the electronic medical record. The measured complications were rippling, defined as none, mild, moderate, or severe; capsular contracture with associated Baker grade; implant exposure; implant rupture; implant infection; hematoma; seroma with documented aspiration; and cellulitis, defined by documented infection requiring antibiotics.
BREAST-Q Assessment
Since 2010, the BREAST-Q patient-reported outcome measure has been administered to all patients as part of routine clinical care at MSK. The following domains are included in the BREAST-Q reconstruction module: (1) Satisfaction with Breast, (2) Satisfaction with Outcome, (3) Psychosocial Well-Being, (4) Physical Well-Being of the Chest and Upper Body, and (5) Sexual Well-Being. In our analysis, we only included patients who completed the BREAST-Q reconstruction module at a minimum of one time point postoperatively. The measured time points were relative to the number of postoperative days following the exchange from tissue expander to breast implant: 3 months, 6 months, 1 year, and 2 years. We used Q-Score software to convert values for BREAST-Q subscales to summary scores ranging from zero to 100, with higher scores correlating with superior outcomes. A difference of four points on the Q-Score was considered to be clinically significant.(26) Satisfaction and well-being scores were the primary domains of interest.
Statistical Analysis
To compare baseline demographic variables between the smooth and textured implant groups, we used the Student’s t-test (continuous variables) or chi-square or Fisher’s exact test (categorical variables). To compare differences in mean (standard deviation [SD]) and median (interquartile range [IQR]) scores for all five domains of the BREAST-Q patient-reported outcome measure between the smooth and textured implant groups, we used the Mann-Whitney test. We used chi-square or Fisher’s exact test to analyze complications. Based on our sample size and an estimated complication effect size (Cohen’s d = 0.2), this study was designed to have 92% power to detect differences in complications, with a type I error rate of 0.05 (two-sided). Additionally, we performed a multivariate linear regression analysis to control for confounding variables (device type, race/ethnicity, BMI, malignancy type, any radiation, any chemotherapy, and reconstruction laterality) with the main outcome of interest being the satisfaction with breast score at the 2-year timepoint. Additionally, we performed a multivariate logistic regression to assess the odds of having any complication, reoperation or revision, controlling for device type, race/ethnicity, BMI, any radiation, any chemotherapy, and reconstruction laterality. All tests were considered significant if the p value was <0.05. We used R statistical software (packages: tidyverse) for statistical analyses.
RESULTS
Demographic and Cancer-Related Characteristics
For our final analysis, we included a total of 1,077 patients who underwent immediate postmastectomy implant-based reconstruction: 785 with smooth implants and 292 with textured implants (Table 1). The mean patient age was 48.2 (9.8) years. The majority of patients were white (928 [86.2%]), never smokers (667 [61.9%]), and had an average BMI of 25.3 (5.0) kg/m2. Most patients presented with a localized tumor (977 [90.7%]), had no radiotherapy (824 [76.5%]), and had either neoadjuvant chemotherapy (471 [43.7%]) or no chemotherapy (602 [55.9%]). Nearly three-quarters of the cohort had bilateral reconstruction (794 [73.7%]).
Table 1.
Patient Demographics and Cancer-Related Characteristics
| Overall (n = 1077) |
Smooth Implant (n = 785) |
Textured Implant (n = 292) |
p | |
|---|---|---|---|---|
| Age, mean years (SD) | 48.2 (9.8) | 48.1 (10.1) | 48.5 (9.1) | 0.622 |
| Race/Ethnicity, n (%) | 0.024 | |||
| White, Hispanic or Non-Hispanic | 928 (86.2) | 665 (84.7) | 263 (90.1) | |
| Black, Hispanic or Non-Hispanic | 68 (6.3) | 55 (7.0) | 13 (4.5) | |
| Asian | 45 (4.2) | 32 (4.1) | 13 (4.5) | |
| Other/Unknown | 36 (3.3) | 33 (4.2) | 3 (1.0) | |
| Smoking Status, n (%) | 0.324 | |||
| Never | 667 (61.9) | 484 (61.7) | 183 (62.7) | |
| Former | 312 (29.0) | 222 (28.3) | 90 (30.8) | |
| Current | 61 (5.7) | 49 (6.2) | 12 (4.1) | |
| Unknown | 37 (3.4) | 30 (3.8) | 7 (2.4) | |
| Hypertension, n (%) | 0.826 | |||
| Yes | 241 (22.4) | 177 (22.6) | 64 (21.9) | |
| No | 836 (77.6) | 608 (77.5) | 228 (78.1) | |
| Diabetes, n (%) | 0.606 | |||
| Yes | 54 (5.0) | 41 (5.2) | 13 (4.5) | |
| No | 1023 (95.0) | 744 (94.8) | 279 (95.6) | |
| BMI, mean kg/m2 (SD) | 25.3 (5.0) | 25.6 (5.3) | 24.4 (4.1) | <0.001 |
| Malignancy History, n (%) | 0.055 | |||
| None | 90 (8.4) | 68 (8.7) | 22 (7.5) | |
| Localized Tumor | 977 (90.7) | 713 (90.8) | 264 (90.4) | |
| Metastatic | 10 (0.9) | 4 (0.5) | 6 (2.1) | |
| Radiotherapy, n (%) | <0.001 | |||
| None | 824 (76.5) | 595 (75.8) | 229 (78.4) | |
| Preoperative (Pre-TE) | 83 (7.7) | 50 (6.4) | 33 (11.3) | |
| Postoperative (Post-TE and Post-Implant) | 170 (15.8) | 140 (17.8) | 30 (10.3) | |
| Chemotherapy, n (%) | 0.374 | |||
| Neoadjuvant | 471 (43.7) | 345 (44) | 126 (43.2) | |
| Adjuvant | 3 (0.3) | 3 (0.4) | 0 (0) | |
| Both | 1 (0.1) | 0 (0) | 1 (0.3) | |
| None | 602 (55.9) | 437 (55.7) | 165 (56.5) | |
| Laterality of Reconstruction, n (%) | 0.149 | |||
| Unilateral | 283 (26.3) | 197 (25.1) | 86 (29.5) | |
| Bilateral | 794 (73.7) | 588 (74.9) | 206 (70.6) |
Abbreviations: BMI body mass index, SD Standard Deviation, TE tissue expander,
p value calculated with a Student's t-test (continuous variables) and with Chi-Square or Fisher's Exact test (categorical variables)
Smooth versus Textured Implant Patients
There were few statistically significant differences between the smooth and textured implant groups, except for race, BMI, and radiotherapy (Table 1). Patients who underwent reconstruction with textured implants were more likely than smooth implant recipients to be white (90.1% vs. 84.7%; p= 0.024) and have a slightly lower average BMI (24.4 vs. 25.6; p<0.001) and a lower rate of postoperative radiotherapy (10.3% vs. 17.8%; p<0.001).
BREAST-Q Scores – Satisfaction
There were no significant differences in satisfaction (Satisfaction with Breast and Satisfaction with Outcome) BREAST-Q scores between smooth and textured implant recipients at any postoperative time point (Table 2, Figure 1). When specifically comparing Satisfaction with Breast scores between non-irradiated patients in the two groups, the only significant difference was a marginal improvement in the smooth implant group at 6 months postoperatively (data not shown).
Table 2.
Patient-Reported BREAST-Q Satisfaction Scores
| Implant Type | Satisfaction with Breast | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| 3 months | p | 6 months | p | 1 year | p | 2 years | p | ||
| Smooth | n | 277 | 0.268 | 268 | 0.163 | 459 | 0.482 | 329 | 0.849 |
| Mean (SD) | 67.9 (16.5) | 67.7 (18.1) | 64.9 (19.9) | 64.6 (19.3) | |||||
| Median (IQR) | 69 (55–78) | 69 (55–78) | 65 (54–78) | 65 (54–78) | |||||
| Textured | n | 59 | 60 | 77 | 101 | ||||
| Mean (SD) | 65.0 (17.4) | 63.9 (17.9) | 66.9 (18.1) | 64.1 (19.4) | |||||
| Median (IQR) | 65 (54.5–74) | 64 (52.8–73.5) | 67 (57–78) | 65 (52–75) | |||||
| Implant Type | Satisfaction with Outcome | ||||||||
| 3 months | p | 6 months | p | 1 year | p | 2 years | p | ||
| Smooth | n | 276 | 0.365 | 269 | 0.702 | 460 | 0.366 | 327 | 0.369 |
| Mean (SD) | 73.6 (20.2) | 73.2 (20.5) | 71.6 (21.7) | 72.2 (21.7) | |||||
| Median (IQR) | 75 (61–100) | 75 (61–100) | 75 (55–89.5) | 75 (61–86) | |||||
| Textured | n | 58 | 60 | 77 | 101 | ||||
| Mean (SD) | 70.8 (21.7) | 71.8 (21.4) | 73.8 (20.0) | 69.3 (24.2) | |||||
| Median (IQR) | 75 (55–86) | 75 (59.3–86) | 75 (67–86) | 75 (51–86) | |||||
Abbreviations: n Count, SD Standard Deviation, IQR Interquartile Range
p value calculated using Mann-Whitney Test
Figure 1: BREAST-Q Satisfaction Outcomes.
Comparison of BREAST-Q satisfaction with breast and outcome scores between patients with smooth versus textured breast implant reconstruction. Means measured at 3 months, 6 months, 1 and 2 years. No statistical significance was observed. (p<0.05).
BREAST-Q Scores – Well-Being
There were no significant differences in Psychosocial Well-Being scores and Physical Well-Being of the Chest scores between the two groups at any postoperative time point (Table 3). Apart from a significant difference in Sexual Well-Being scores at 1 year (smooth implant group: mean = 52.3 [SD =22.1]; textured implant group: mean = 58.6 [22.5]; p = 0.039; Table 3), there were no other statistical differences in Sexual Well-Being scores. Because of the higher proportion of radiated patients in the smooth implant group, a stratified analysis was performed isolating the effects of XRT. A subgroup analysis comparing BREAST-Q satisfaction and well-being scores in non-irradiated patients demonstrated no significant difference between the smooth and textured implant groups (data not shown).
Table 3.
Patient-Reported BREAST-Q Well-Being Scores
| Implant Type | Psychosocial Well-Being | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| 3 months | p | 6 months | p | 1 year | p | 2 years | p | ||
| Smooth | n | 277 | 0.363 | 269 | 0.214 | 460 | 0.194 | 328 | 0.821 |
| Mean (SD) | 74.0 (19.7) | 73.4 (20.1) | 73.3 (20.6) | 74.8 (20.7) | |||||
| Median (IQR) | 76 (58–92) | 70 (58–92) | 73 (58–92) | 76 (60–92) | |||||
| Textured | n | 58 | 60 | 77 | 101 | ||||
| Mean (SD) | 71.2 (21.1) | 76.3 (22.0) | 76.8 (19.6) | 75.3 (20.5) | |||||
| Median (IQR) | 73 (55.5–86) | 76 (63–100) | 79 (60–100) | 76 (63–92) | |||||
| Implant Type | Physical Well-Being of the Chest | ||||||||
| 3 months | p | 6 months | p | 1 year | p | 2 years | p | ||
| Smooth | n | 277 | 0.478 | 269 | 0.597 | 460 | 0.088 | 327 | 0.194 |
| Mean (SD) | 72.3 (14.7) | 73.9 (15.7) | 73.7 (15.3) | 76.5 (16.2) | |||||
| Median (IQR) | 71 (63–81) | 74 (63–85) | 74 (63–85) | 77 (66–85) | |||||
| Textured | n | 58 | 60 | 77 | 101 | ||||
| Mean (SD) | 70.3 (15.4) | 72.7 (14.5) | 76.9 (15.2) | 74.1 (16.8) | |||||
| Median (IQR) | 71 (60–81) | 71 (63–81) | 74 (68–85) | 74 (66–85) | |||||
| Implant Type | Sexual Well-Being | ||||||||
| 3 months | p | 6 months | p | 1 year | p | 2 years | p | ||
| Smooth | n | 265 | 0.355 | 258 | 0.457 | 441 | 0.039 | 317 | 0.781 |
| Mean (SD) | 54.8 (22.9) | 54.2 (21.8) | 52.3 (22.1) | 55.4 (23.8) | |||||
| Median (IQR) | 54 (43–63) | 53 (39–67) | 51 (39–63) | 52 (42–67) | |||||
| Textured | n | 58 | 59 | 75 | 98 | ||||
| Mean (SD) | 51.4 (21.2) | 56.1 (21.1) | 58.6 (22.5) | 55.5 (21.3) | |||||
| Median (IQR) | 52 (37.5–63) | 54 (41–67) | 57 (44–74.5) | 56 (43–63) | |||||
Abbreviations: n Count, SD Standard Deviation, IQR Interquartile Range
p value calculated using Mann-Whitney Test
Regression analysis: BREAST-Q Scores
To control for differences between the implant cohorts as well as factors understood to be related to satisfaction with breast scores, we performed a multivariate linear regression analysis with the primary outcome of interest being satisfaction with breasts at the 2-year timepoint (Table 4). Controlling for these variables, surface texture type was not significantly associated with satisfaction with breasts.
Table 4:
Multivariate Linear Regression Analysis of 2-year Satisfaction with Breast Scores
| β Coefficient | Standard Error | p value | |
|---|---|---|---|
| Device Type | |||
| Smooth | (ref) | ||
| Textured | −1.2 | 2.1 | 0.57 |
| Race/Ethnicity | |||
| White, Hispanic or Non-Hispanic | (ref) | ||
| Black, Hispanic or Non-Hispanic | 1.8 | 3.8 | 0.63 |
| Asian | −0.8 | 4.5 | 0.86 |
| Other/Unknown | −4.2 | 4.7 | 0.37 |
| Malignancy Type | |||
| None | (ref) | ||
| Localized Tumor | −1.5 | 3.5 | 0.66 |
| Metastatic | −0.8 | 13.6 | 0.95 |
| BMI | −0.3 | 0.2 | 0.07 |
| Any Radiation | |||
| None | (ref) | ||
| Yes | −7.9 | 2.3 | <0.001 |
| Any Chemotherapy | |||
| None | (ref) | ||
| Yes | −3.2 | 2.0 | 0.11 |
| Reconstruction Laterality | |||
| Bilateral | (ref) | ||
| Unilateral | −7.6 | 2.1 | <0.001 |
| Intercept | 79.6 | 5.3 | < 0.001 |
Abbreviations: ref Reference
Linear regression assessing BREAST-Q Satisfaction with Breast Scores
Complications
Overall, there was a higher rate of any complication, revision or re-operation in the textured implant group (130 [44.5%]) compared to the smooth implant group (276 [35.2%]; p < 0.001; Table 5). There was a significant difference in the incidence of moderate and severe rippling between the textured and smooth implant groups where patients with smooth implants had a higher rate of moderate and severe rippling compared to textured implant patients (p = 0.003). Conversely, patients who underwent reconstruction with textured implants had a significantly higher rate of cellulitis than patients who underwent reconstruction with smooth implants (smooth implant group: 41 [5.2%]; textured implant group: 27 [9.2%]; p = 0.016). All other complications did not differ statistically between the two groups, including rates of capsular contracture by grade, implant exposure, implant leak/rupture, or infected implant. Total length of follow up was significantly longer for the textured patient cohort (p<0.001).
Table 5:
Postoperative Complication Rates
| Overall (n = 1077) |
Smooth Implant (n = 785) |
Textured Implant (n = 292) |
p value | |
|---|---|---|---|---|
| Any complication, revision or re-operation, n (%) | 406 (37.7) | 276 (35.2) | 130 (44.5) | 0.005 |
| Implant Exchange, n (%) | 201 (18.7) | 117 (14.9) | 84 (28.8) | < 0.001 |
| Implant Removal, n (%) | 16 (1.5) | 10 (1.3) | 6 (2.1) | 0.395 |
| Implant Conversion to Autologous (Implant failure), n (%) | 22 (2.0) | 14 (1.8) | 8 (2.7) | 0.324 |
| Implant Rippling (moderate and severe), n (%) | 58 (5.4) | 52 (6.6) | 6 (2.1) | 0.003 |
| Capsular Contracture (Grade 3 and 4), n (%) | 132 (12.3) | 92 (11.7) | 40 (13.7) | 0.379 |
| Exposed Implant, n (%) | 4 (0.4) | 2 (0.3) | 2 (0.7) | 0.298 |
| Implant Rupture, n (%) | 19 (1.8) | 12 (1.5) | 7 (2.4) | 0.336 |
| Infected Implant, n (%) | 5 (0.5) | 4 (0.5) | 1 (0.3) | 1 |
| Hematoma, n (%) | 23 (2.1) | 19 (2.4) | 4 (1.4) | 0.289 |
| Seroma, n (%) | 40 (3.7) | 32 (4.1) | 8 (2.7) | 0.302 |
| Cellulitis, n (%) | 68 (6.3) | 41 (5.2) | 27 (9.2) | 0.016 |
| Duration of Provider Follow Up * | <0.001 | |||
| Mean days (SD) | 2042 (990.5) | 1796.9 (865.5) | 2705.8 (1004.2) | |
| Median days (IQR) | 1966 (1246 – 2713) | 1710 (1122 – 2365) | 2944.5(2057.3 – 3477.5) |
p value calculated with Chi-Square or Fisher's Exact test
Follow up time p value calculated using Mann-Whitney test as distribution of time was not normal
Regression Analysis: Any complication, revision or re-operation
A multivariate logistic regression was performed (Table 6) to control for factors associated with any complication, revision or re-operation. Textured implants had a significantly higher odds of any complication, revision or re-operation compared to smooth implants, keeping all other variables constant (OR = 1.6; 95% CI = 1.2 – 2.1; p=0.002). As well, patients with any radiation had significantly higher odds of any complication revision or re-operation compared to no radiation (OR = 2.2; 95% CI = 1.6 – 2.9; p < 0.001). For diagnostic purposes, we examined this model with a likelihood ratio test. The model was a better fit of the data than an intercept only model or a model that only included type of implant (p<0.001).
Table 6:
Multivariate Logistic Regression of Any complication, revision or re-operation
| OR | 95% CI, Lower Limit |
95% CI, Upper Limit |
p value | |
|---|---|---|---|---|
| Device Type | ||||
| Smooth | (ref) | |||
| Textured | 1.6 | 1.2 | 2.1 | 0.002 |
| Race/Ethnicity | ||||
| White, Hispanic or Non-Hispanic | (ref) | |||
| Black, Hispanic or Non-Hispanic | 1.0 | 0.6 | 1.7 | 0.97 |
| Asian | 1.1 | 0.6 | 2.0 | 0.83 |
| Other/Unknown | 1.3 | 0.6 | 2.5 | 0.53 |
| BMI | 1.0 | 1.0 | 1.0 | 0.99 |
| Any Radiation | ||||
| None | (ref) | |||
| Yes | 2.2 | 1.6 | 2.9 | <0.001 |
| Any Chemotherapy | ||||
| None | (ref) | |||
| Yes | 1.3 | 1.0 | 1.6 | 0.1 |
| Reconstruction Laterality | ||||
| Bilateral | (ref) | |||
| Unilateral | 0.8 | 0.6 | 1.1 | 0.12 |
Abbreviations: ref Reference, OR Odds Ratio, 95% CI 95% Confidence Interval
Logistic regression assessing odds of any complication, revision or re-operation adjusting for device type, race/ethnicity, BMI, any radiation, any chemotherapy, and reconstruction laterality
DISCUSSION
In our analysis of over 1000 patients who underwent immediate postmastectomy implant-based breast reconstruction, patient-reported satisfaction and psychosocial and physical well-being did not differ significantly between implant surface type: smooth versus textured. Patient-reported sexual well-being did differ significantly between smooth and textured implant recipients but only at 1 year and not at any of the other time points (3 months, 6 months, or 2 years). Thus, although the patients in this study did not transition from a textured to a smooth implant, our results can be used to inform patients that both satisfaction and quality of life following postmastectomy breast reconstruction appear to be independent of implant surface type.
The merits of smooth versus textured breast implants have been the subject of numerous studies, with mixed results. Although textured implants are generally thought to be associated with less capsular contracture, lower re-operation rate, less implant malposition, and less rippling than smooth implants,(9-14) other studies have not demonstrated improved surgical or patient-reported outcomes.(18, 22, 27) These heterogenous results have left patients and plastic surgeons with questions regarding the true extent to which the surface texture of an implant affects outcomes in breast reconstruction. Given the ongoing focus on BIA-ALCL, the discussion has shifted away slightly from the merits of textured devices to the complication profile of these implants. Many patients who have textured devices now present to discuss replacing their textured implants with smooth surface devices. The results of our study can assist with the perioperative counseling of these patients.
Historically, textured breast implants have been linked to decreased rates of capsular contracture.(20, 21) In our study, we noted comparable degrees of capsular contracture rates between the smooth and textured implant groups. Our finding is in line with an increasing number of studies reporting that capsular contracture is independent of implant texture and challenging the idea of the protective effect of textured implants on capsular contracture.(9, 14, 28)
While the complication profiles of the smooth and textured implant groups in our study were largely similar, there were some differences. We found a significant increase in rippling in smooth implant recipients, but overall rates of rippling were consistent with those in recently published studies.(29, 30) This finding was not surprising given that smooth implants tend to contain gel of lower cohesivity, while textured implants, particularly anatomic ones, have more form-stable gels.(31) However, newer generations of smooth implants are filled with more cohesive gels, which could minimize the difference in rippling between the two implant surface types.(10, 11) Additionally, fat grafting has become increasingly common as a means to help mask rippling, although the current study did not directly assess this technique. We also noted that a higher proportion of textured implant recipients in our cohort experienced cellulitis, compared with the smooth implant group. This may be related to the increased adherence of bacteria and formation of biofilm on textured breast implants.(17, 32) Additionally, regression analysis confirmed higher odds of any complication, revision or re-operation overall based on surface type. Our current findings are in contrast with those from an earlier report from this institution on 365 patients that showed no significant difference in complications based on implant surface.(18) These differing results may be explained by the current study’s larger population and longer follow-up. However, it is important to note that follow up time was significantly longer for the textured devices, which biases complications, revisions or re-operations towards the textured group given longer device exposure.
A major strength of our study is that it is the largest to compare patient-reported outcomes between smooth and textured breast implant recipients. Despite our large cohort, however, our study remains underpowered to measure small differences in BREAST-Q outcomes. There are also other limitations that merit comment. The study may suffer from a healthy responder bias, as it is possible that patients who were happier with their results were more likely to complete the BREAST-Q assessment. Additionally, our group only recently began to consistently collect pre-operative BREAST-Q data and as a result we were unable to compare the two groups pre-operatively in the study. There is also a difference in the BREAST-Q response rate between the two groups and by postoperative time point. The BREAST-Q response rate was lower among textured implant recipients than among those with smooth implants. This difference in response rate between groups may be largely attributed to the greater number of patients with smooth implants in the study, paralleling the increase in the use of smooth breast implants over textured devices and improvements in the administration of BREAST-Q over the years. The noted difference in BREAST-Q response rate by postoperative time point is an active area of quality improvement in our institution.
Another limitation of our study is the significantly different proportion of irradiated patients in each implant group. To adjust for this difference, we evaluated BREAST-Q scores from non-irradiated patients in the two implant groups and found them to be not statistically different at all time points. These results suggest that in the absence of radiotherapy, patient-reported outcomes are independent of implant texture. Sample size limitations in the radiation cohort did not allow for meaningful interpretation of the subgroup results.
Lastly, complications were determined through chart review, which may not capture all events, given the potential for observer and initial physician recorder bias. This method of data collection requires that the treating physician note any physical exam findings of interest (such as rippling). These is a possibility of selection bias in that some differences between cohorts were noted. Further, the length of follow up time was significantly longer for textured patients, reflecting utilization use differences over time. This creates a bias in the complication analysis, as complication rates were not adjusted for length of follow up and the results should be viewed considering this. While a strength of our study was that it was sufficiently powered to measure complications, the sample had limited ethnic and racial diversity, with the majority being white and non-Hispanic. Patients were from a single, urban, high-volume academic center, so findings may not be generalizable to women treated at smaller, nonacademic centers. However, the same group of attending surgeons treated both cohorts over the study period all of whom utilized smooth and textured devices; therefore system-level issues and biases would be equal between groups. To note, for the multivariate logistic regression assessing odds for complications, revisions and re-operations, the odds ratio may overestimate the true risk ratio of major complications between groups as this aggregate outcome was not rare in this cohort.
CONCLUSIONS
The surface texture of breast implants may not directly impact patient-reported outcomes following immediate postmastectomy breast reconstruction. The current findings can be used to inform patients that despite the purported benefits of textured devices, their outcomes will not be compromised with the use of smooth devices. The results will also aid in the informed consent process for patients considering an exchange from textured to smooth devices. Patient-reported outcomes of implant exchange procedures remains an area for future research.
Source of Funding:
This research was funded in part through the NIH/NCI Cancer Center Support Grant P30CA008748.
Footnotes
Financial Disclosure Statement: Andrea Pusic is a codeveloper of the BREAST-Q, which is owned by Memorial Sloan Kettering Cancer Center. She receives a portion of the licensing fees (royalty payments) when the BREAST-Q is used in industry-sponsored clinical trials. Joseph Dayan is a consultant for Stryker. Babak Mehrara is a consultant for PureTech Corporation. The remaining authors declare no conflicts of interest.
REFERENCES
- 1.ASPS. American Society of Plastic Surgeons. 2018 Plastic Surgery Statistics Report. 2019.
- 2.Clemens MW, Miranda RN Coming of age: breast implant-associated anaplastic large cell lymphoma after 18 years of investigation. Clin Plast Surg 42: 605–613, 2015. [DOI] [PubMed] [Google Scholar]
- 3.US Food and Drug Administration. Allergan Recalls Natrelle Biocell Textured Breast Implants Due to Risk of BIA-ALCL Cancer.
- 4.Carr LW, Roberts J, Potochny JD How breast implant surface type is influenced by breast implant-associated anaplastic large cell lymphoma: a survey of the American Society of Plastic Surgeons. Ann Plast Surg 82: S208–S211, 2019. [DOI] [PubMed] [Google Scholar]
- 5.Tandon VJ, DeLong MR, Ballard TN, et al. Evolving trends in textured implant use for cosmetic augmentation in the United States. Plast Reconstr Surg 142: 1456–1461, 2018. [DOI] [PubMed] [Google Scholar]
- 6.US Food and Drug Administration. The FDA Requests Allergan Voluntarily Recall Natrelle BIOCELL Textured Breast Implants and Tissue Expanders from the Market to Protect Patients: FDA Safety Communication.
- 7.McGuire PA, Deva AK, Glicksman CA, et al. Management of asymptomatic patients with textured surface breast implants. Aesthetic Surgery Journal Open Forum 1, 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Lapid O, Noels EC, Meijer SL Pathologic findings in primary capsulectomy specimens: analysis of 2531 patients. Aesthet Surg J 34: 714–718, 2014. [DOI] [PubMed] [Google Scholar]
- 9.Barnsley GP, Sigurdson LJ, Barnsley SE Textured surface breast implants in the prevention of capsular contracture among breast augmentation patients: a meta-analysis of randomized controlled trials. Plast Reconstr Surg 117: 2182–2190, 2006. [DOI] [PubMed] [Google Scholar]
- 10.Calobrace MB, Capizzi PJ The biology and evolution of cohesive gel and shaped implants. Plast Reconstr Surg 134: 6S–11S, 2014. [DOI] [PubMed] [Google Scholar]
- 11.Calobrace MB, Schwartz MR, Zeidler KR, et al. Long-term safety of textured and smooth breast implants. Aesthet Surg J 38: 38–48, 2017. [DOI] [PubMed] [Google Scholar]
- 12.Maxwell GP, Scheflan M, Spear S, et al. Benefits and limitations of macrotextured breast implants and consensus recommendations for optimizing their effectiveness. Aesthet Surg J 34: 876–881, 2014. [DOI] [PubMed] [Google Scholar]
- 13.Maxwell GP, Van Natta BW, Bengtson BP, et al. Ten-year results from the Natrelle 410 anatomical form-stable silicone breast implant core study. Aesthet Surg J 35: 145–155, 2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Somogyi RB, Brown MH Outcomes in primary breast augmentation: a single surgeon's review of 1539 consecutive cases. Plast Reconstr Surg 135: 87–97, 2015. [DOI] [PubMed] [Google Scholar]
- 15.Adams WP Jr., Mallucci P Breast augmentation. Plast Reconstr Surg 130: 597e–611e, 2012. [DOI] [PubMed] [Google Scholar]
- 16.Mallucci P Discussion: Intraoperative comparison of anatomical versus round implants in breast augmentation: a randomized controlled trial. Plast Reconstr Surg 139: 599–600, 2017. [DOI] [PubMed] [Google Scholar]
- 17.Adams WP Jr., Culbertson EJ, Deva AK, et al. Macrotextured breast implants with defined steps to minimize bacterial contamination around the device: experience in 42,000 implants. Plast Reconstr Surg 140: 427–431, 2017. [DOI] [PubMed] [Google Scholar]
- 18.Antony AK, McCarthy C, Disa JJ, et al. Bilateral implant breast reconstruction: outcomes, predictors, and matched cohort analysis in 730 2-stage breast reconstructions over 10 years. Ann Plast Surg 72: 625–630, 2014. [DOI] [PubMed] [Google Scholar]
- 19.Shauly O, Gould DJ, Patel KM Microtexture and the cell/biomaterial interface: a systematic review and meta-analysis of capsular contracture and prosthetic breast implants. Aesthet Surg J 39: 603–614, 2019. [DOI] [PubMed] [Google Scholar]
- 20.Hakelius L, Ohlsen L A clinical comparison of the tendency to capsular contracture between smooth and textured gel-filled silicone mammary implants. Plast Reconstr Surg 90: 247–254, 1992. [PubMed] [Google Scholar]
- 21.Pollock H Breast capsular contracture: a retrospective study of textured versus smooth silicone implants. Plast Reconstr Surg 91: 404–407, 1993. [PubMed] [Google Scholar]
- 22.Macadam SA, Ho AL, Lennox PA, et al. Patient-reported satisfaction and health-related quality of life following breast reconstruction: a comparison of shaped cohesive gel and round cohesive gel implant recipients. Plast Reconstr Surg 131: 431–441, 2013. [DOI] [PubMed] [Google Scholar]
- 23.Agochukwu-Nwubah N, Boustany A, Wetzel M, et al. Anatomic Implants in Breast Reconstruction: A Comparison of Outcomes and Aesthetic Results Compared to Smooth Round Silicone Implants. Aesthet Surg J 39: NP322–NP330, 2019. [DOI] [PubMed] [Google Scholar]
- 24.Buonomo OC, Morando L, Materazzo M, et al. Comparison of round smooth and shaped micro-textured implants in terms of quality of life and aesthetic outcomes in women undergoing breast reconstruction: a single-centre prospective study. Updates Surg, 2020. [DOI] [PubMed] [Google Scholar]
- 25.Khavanin N, Clemens MW, Pusic AL, et al. Shaped versus Round Implants in Breast Reconstruction: A Multi-Institutional Comparison of Surgical and Patient-Reported Outcomes. Plast Reconstr Surg 139: 1063–1070, 2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Voineskos SH, Klassen AF, Cano SJ, et al. Giving meaning to differences in BREAST-Q scores: minimal important difference for breast reconstruction patients. Plast Reconstr Surg 145: 11e–20e, 2020. [DOI] [PubMed] [Google Scholar]
- 27.Macadam SA, Ho AL, Cook EF Jr., et al. Patient satisfaction and health-related quality of life following breast reconstruction: patient-reported outcomes among saline and silicone implant recipients. Plast Reconstr Surg 125: 761–771, 2010. [DOI] [PubMed] [Google Scholar]
- 28.Calobrace MB, Stevens WG, Capizzi PJ, et al. Risk factor analysis for capsular contracture: a 10-year Sientra study using round, smooth, and textured implants for breast augmentation. Plast Reconstr Surg 141: 20S–28S, 2018. [DOI] [PubMed] [Google Scholar]
- 29.Pantelides NM, Srinivasan JR Rippling following breast augmentation or reconstruction: aetiology, emerging treatment options and a novel classification of severity. Aesthetic Plast Surg 42: 980–985, 2018. [DOI] [PubMed] [Google Scholar]
- 30.Sigalove S, Maxwell GP, Gabriel A Outcomes utilizing Inspira implants in revisionary reconstructive surgery. Plast Reconstr Surg 144: 66S–72S, 2019. [DOI] [PubMed] [Google Scholar]
- 31.Panettiere P, Marchetti L, Accorsi D Soft cohesive silicone gel breast prostheses: a comparative prospective study of aesthetic results versus lower cohesivity silicone gel prostheses. J Plast Reconstr Aesthet Surg 60: 482–489, 2007. [DOI] [PubMed] [Google Scholar]
- 32.Jones P, Mempin M, Hu H, et al. The functional influence of breast implant outer shell morphology on bacterial attachment and growth. Plast Reconstr Surg 142: 837–849, 2018. [DOI] [PubMed] [Google Scholar]