Abstract
Background
The growing use of tranexamic acid (TXA) in plastic surgery warrants an accurate analysis of efficacy for body contouring indications. We offer the largest cohort evaluation of intravenous TXA use in minimizing postoperative complications following reduction mammoplasty.
Methods
A single-institution retrospective analysis was conducted on consecutive reduction mammoplasty procedures performed between February 2023–August 2024. Patient demographics, intraoperative factors, and postoperative complications were collected and summarized using descriptive statistics. Groups were stratified based on administration of TXA, with differences assessed using Pearson’s chi-square test for categorical variables and Student’s t-test or Wilcoxon rank sum test for continuous variables, as appropriate. Association of perioperative intravenous TXA use with postoperative complications was analyzed by breast with multivariable logistic regression.
Results
Of 406 patients (812 breasts), 213 patients (426 breasts) were administered TXA perioperatively. Neither group had significant differences in Caprini score, prior history of DVT, or anemia. Multivariable regression analysis, controlling for differences in demographics and operative characteristics, demonstrated that TXA use was significantly associated with decreased rates of dehiscence (aOR = 0.50, 95% CI 0.28–0.91, p = 0.023) and scar revision (aOR = 0.30, 95% CI 0.12–0.76, p = 0.015). Among breasts that were only closed with barbed sutures, after adjustment for demographics and operative characteristics, TXA use was significantly associated with lower rates of scar revision only (aOR = 0.16; 95% CI 0.05–0.52; p = 0.005).
Conclusion
Intravenous TXA use in reduction mammoplasty confers lower rates of postoperative wound complications, though its impact may be secondary to wound closure technique.
Keywords: Outcomes studies, Retrospective studies, Reduction mammoplasty, Tranexamic acid, Wound dehiscence
Introduction
Reduction mammoplasty is a common plastic surgery procedure, widely regarded as safe and reproducible, yielding significant improvement in patient-reported outcomes. However, some of the feared adverse events that can occur include hematoma and subsequent readmission for evacuation.1 Even smaller-sized hematomas can lead to wound-healing complications, such as dehiscence or poor wound healing, thereby contributing to subpar postoperative outcomes.
To minimize hematoma rates, surgeons have begun experimenting with the intraoperative use of topical and intravenous (IV) tranexamic acid (TXA).2 Studies from other surgical specialties, such as orthopedics, cardiothoracic, and neurosurgery, have demonstrated that TXA use reduces intraoperative bleeding and decreases the need for subsequent blood transfusions.3,4 Although transfusions are unlikely to be a consequence of reduction mammoplasties and most body contouring procedures, plastic surgeons have found use for TXA during breast or burn reconstruction and craniofacial procedures, such as rhytidectomy or rhinoplasty, to minimize hematoma rates.2,5 Despite the strong evidence for the efficacy of TXA in plastic surgery procedures, few studies have demonstrated its effectiveness in reduction mammoplasty.
As reduction mammoplasty is a standard plastic surgery procedure, some studies analyzing TXA use in breast surgery have aimed to expand their cohorts to include these cases. A few randomized controlled trials have shown that TXA use can decrease hematoma rates in reduction mammoplasty; however, they rarely analyze IV TXA over topical dosing.6 Orthopedic studies have compared IV and topical TXA across various arthroplasty procedures, finding that both can reduce transfusions and intraoperative blood loss at comparable rates.7,8 Despite this, it is evident that IV TXA has the added benefit of ease of administration without increasing the time required to apply the topical form. Generally, the plastic surgery literature has limited data on topical versus IV TXA.
Our study, therefore, analyzed the effect of IV TXA use specifically on reduction mammoplasty outcomes by evaluating the largest dataset undergoing breast reduction with concurrent IV TXA administration. Based on previous literature and our institution’s practice, we hypothesized that IV TXA use would minimize postoperative complications following reduction mammoplasty.
Methods
Data source and collection
An Institutional Review Board (IRB) approved retrospective-designed study was conducted on 406 patients (812 breasts) who underwent bilateral reduction mammoplasties by six senior surgeons at our institution from February 2023 to August 2024. The decision to use TXA, the timing of its administration, and additional intraoperative choices (pedicle type, barbed suture usage, and closed-suction drain placement) were based on surgeon preferences for each operation. Patients who underwent reductions for gynecomastia and those receiving concurrent liposuction with adrenaline-based tumescent solution (± topical TXA) were excluded. All patients had Wise-pattern skin excisions performed for their reductions and were seen for at least one postoperative follow-up appointment. Patient demographics, including age, gender, preoperative body mass index (BMI), smoking history, prior breast surgeries, preoperative anticoagulant use, Caprini score, anemia, and prior deep vein thrombosis (DVT), were extracted from the medical record. Additional intraoperative data collected included the weight removed per breast, use of barbed suture for closure, presence of postoperative drains, duration of procedure, estimated blood loss, and choice of pedicle. Post-surgical complications, such as breast seroma, hematoma, wound dehiscence, infection, nipple insensitivity, hypertrophic scarring/keloid, need for scar revision, and fat necrosis, were also extracted from the medical records.
Surgical techniques
All patients were marked in the preoperative area where the physician could note any asymmetries in breast size while they were sitting upright or standing. Once in the operating room, some patients (depending on surgeon preference) were administered 1 g of IV TXA. Most patients in the TXA cohort received the antifibrinolytic during anesthesia induction or within 30 min of the procedure start time. Only 6 of the 213 patients who received TXA were administered the agent >30 min after the procedure start time.
A Wise-pattern skin reduction with an inferior, medial, superomedial, or inferomedial dermoglandular pedicle was designed, and the breast reduction procedure was performed in a standard fashion as previously described.9 At the end of the procedure, some patients had drains placed at the surgeon’s discretion. Finally, incisions were closed according to the surgeon’s preference, either in a conventional layered fashion with Monocryl or Vicryl for the dermal layer and Monocryl for subcuticular closure, or with Stratafix, Ethibond, or Quill barbed sutures. Dressings were then applied. Following complete closure of all incisions, local Lidocaine or Marcaine with epinephrine was injected into each breast for postoperative pain control.
Statistical analysis
Patient demographics, comorbidities, and systemic intraoperative characteristics (procedure duration, estimated blood loss, pedicle type) were analyzed by patient. Comparison of weight removed per breast, barbed suture use, closed-suction drain use, and complications were analyzed by breast. TXA and non-TXA cohorts of patients and their breasts were established depending on whether IV TXA was given perioperatively, as recorded in the anesthesia note.
To compare demographics, operative characteristics, and postoperative outcomes between the two cohorts, Student’s t-test or Wilcoxon rank sum test was applied for continuous variables, as appropriate, and Pearson’s chi-square test was used for categorical variables. Univariate and multivariable logistic regression analyses using Firth’s correction to account for low event frequencies were performed to estimate odds ratios (ORs) and adjusted odds ratios (aORs) with 95% confidence intervals, evaluating the association between IV TXA use and wound complications. Multiple logistic regression models were constructed to account separately for patient demographics, operative characteristics, and a combination of variables encompassing both demographics and operative features. An additional regression analysis was performed on all breasts with incisions closed with barbed sutures only, as there was a considerable discrepancy at baseline between the percentages of breasts closed with and without barbed sutures between the TXA cohorts. All listed p-values are two-tailed. Python 3.7 (Python Software Foundation) and R version 4.2.3 (R Core Team 2023) were used for data preprocessing and analysis. A p-value of <0.05 was determined a priori as statistically significant.
Results
Patient baseline demographics and operative characteristics
Of the total 406 patients (812 breasts), 213 (426 breasts) received IV TXA intraoperatively, while 193 (386 breasts) did not. All patients included in the study self-identified as female. Patients averaged an age of 40 ± 15 years, with a preoperative BMI of 31.5 ± 4.6 kg/m², and a Caprini score for venous thromboembolism risk of 3.9 ± 1.2 (Table 1). Only 9 (2.2%) patients were current smokers, while 71 (17.5%) had previously smoked but quit before the surgery. Small percentages of patients had preoperative diagnoses of hypertension (15.3%), diabetes (3.9%), and anemia (7.9%). 6.2% of patients were on anticoagulant therapy before surgery, and 1.2% of patients were previously treated for DVT. In comparing the demographic characteristics of patients who received IV TXA with those who did not, patients in the TXA group had a significantly lower preoperative BMI compared to those who did not receive TXA. No statistically significant differences were found between the two groups among the remaining demographic and preoperative variables.
Table 1.
Demographics and medical history of patients who did and did not receive intraoperative TXA.
| All patients | TXA used | No TXA used | p-value | |
|---|---|---|---|---|
| Number of patients | 406 | 213 | 193 | |
| Age (years) | 40 ± 15 | 40 ± 15 | 39 ± 16 | 0.555 |
| Smoking History | 17.5% (71) | 20.2% (43) | 14.5% (28) | 0.30 |
| Never | 80.3% (326) | 77.9% (166) | 82.9% (160) | 0.258 |
| Former | 17.5% (71) | 20.2% (43) | 14.5% (28) | 0.169 |
| Current | 2.2% (9) | 1.9% (4) | 2.6% (5) | 0.881 |
| Hypertension | 15.3% (62) | 13.1% (28) | 17.6% (34) | 0.266 |
| Diabetes | 3.9% (16) | 2.8% (6) | 5.2% (10) | 0.333 |
| Preoperative anticoagulant use | 6.2% (25) | 4.7% (10) | 7.8% (15) | 0.28 |
| Nutritional deficiency | 1.0% (4) | 0.9% (2) | 1.0% (2) | 0.686 |
| Preoperative BMI (kg/m2) | 31.5 ± 4.6 | 31.0 ± 4.5 | 32.0 ± 4.7 | 0.004 |
| Caprini score | 3.9 ± 1.2 | 3.8 ± 1.0 | 4.0 ± 1.4 | 0.066 |
| History of breast cancer | 0.7% (3) | 0.0% (0) | 1.6% (3) | 0.213 |
| Anemia | 7.9% (32) | 7.5% (16) | 8.3% (16) | 0.915 |
| Prior DVT | 1.2% (5) | 0.9% (2) | 1.6% (3) | 0.912 |
| Prior breast surgery | 12.1% (49) | 9.4% (20) | 15.0% (29) | 0.112 |
| History of weight loss | 13.5% (55) | 10.3% (22) | 17.1% (33) | 0.065 |
Frequencies of categorical variables are reported as % (n).
Continuous variables are reported as mean ± standard deviation.
Boldface indicates statistical significance (p<0.05).
For operative characteristics, estimated blood loss among all patients was 95.0 ± 52.3 mL, with no significant difference between patients who received and did not receive TXA (Table 2). There were significant differences in pedicle choice between the two groups, as well as in procedure duration (p < 0.001). Breasts of patients who received TXA were more likely to have closed-suction drains placed (47.9% vs. 26.9%, p < 0.001) and barbed suture closure (98.1% vs. 36.8%, p < 0.001). In addition, breasts in the TXA cohort had significantly greater weight of tissue removed (772.7 ± 342.2 vs. 649.5 ± 329.5, p < 0.001) but shorter sternal notch to NAC length (31.6 ± 4.1 vs. 32.5 ± 4.5, p = 0.004).
Table 2.
Operative characteristics of patients who did and did not receive intraoperative TXA.
| All patients | TXA used | No TXA used | p-values | |
|---|---|---|---|---|
| Count (Patients) | 406 | 213 | 193 | |
| Procedure duration | 157.9 ± 51.8 | 149.3 ± 42.3 | 167.3 ± 59.2 | <0.001 |
| Combined procedure | 5.9% (24) | 6.1% (13) | 5.7% (11) | 0.969 |
| Estimated blood loss (mL) | 95.0 ± 52.3 | 90.4 ± 53.1 | 100.1 ± 50.9 | 0.062 |
| Pedicle Type | ||||
| Inferior | 38.7% (157) | 26.8% (57) | 51.8% (100) | <0.001 |
| Superomedial | 53.2% (216) | 72.3% (154) | 32.1% (62) | <0.001 |
| Inferomedial | 4.7% (19) | 0.0% (0) | 9.8% (19) | <0.001 |
| Medial | 3.4% (14) | 0.9% (2) | 6.2% (12) | 0.008 |
| Count (Breasts) | 812 | 426 | 386 | |
| Closed-suction drain used | 37.9% (308) | 47.9% (204) | 26.9% (104) | <0.001 |
| Barbed suture usage | 69.0% (560) | 98.1% (418) | 36.8% (142) | <0.001 |
| Sternal notch to NAC (cm) | 32.0 ± 4.3 | 31.6 ± 4.1 | 32.5 ± 4.5 | 0.004 |
| NAC to IMF (cm) | 14.9 ± 3.6 | 15.1 ± 3.0 | 14.7 ± 4.1 | 0.156 |
| Weight removed (by breast) (g) | 739.4 ± 568.8 | 772.7 ± 342.2 | 649.5 ± 329.5 | <0.001 |
Frequencies of categorical variables are reported as % (n).
Continuous variables are reported as mean ± standard deviation.
Boldface indicates statistical significance (p<0.05).
Postoperative complications
The average follow-up of all patients was 141 ± 98 days; there was no statistically significant difference in postoperative days to last follow-up between patients who received IV TXA and those who did not. At baseline, before adjusting for differences between both cohorts, dehiscence and hypertrophic scarring rates were significantly higher among breasts of patients who did not receive TXA (Table 3). A total of 20 patients were readmitted for scar revision surgery, out of which 12 did not receive TXA, and 8 did, yielding no statistically significant difference between the cohorts (p = 0.366). A total of 10 patients were readmitted for surgical intervention, 6 requiring hematoma evacuation (3 received TXA), 3 requiring debridement (1 received TXA) and closure following dehiscence, and 1 requiring abscess drainage and washout. There was no significant difference in readmission for surgical intervention between cohorts (p = 0.632); this lack of statistical significance remained when stratified by intervention required.
Table 3.
Post-operative follow-up time and complications among patients who did and did not receive intraoperative TXA.
| All | TXA used | No TXA used | p-values | |
|---|---|---|---|---|
| PODs to last follow-up (days) | 141 ± 98 | 140 ± 93 | 142 ± 103 | 0.796 |
| Complication by breast | ||||
| Number of breasts | 812 | 426 | 386 | |
| Seroma | 1.4% (11) | 0.7% (3) | 2.1% (8) | 0.165 |
| Hematoma | 2.7% (22) | 1.6% (7) | 3.9% (15) | 0.08 |
| Dehiscence | 9.6% (78) | 4.9% (21) | 14.8% (57) | <0.001 |
| Infection | 1.4% (11) | 1.6% (7) | 1.0% (4) | 0.672 |
| Nipple insensitivity | 5.9% (48) | 6.3% (27) | 5.4% (21) | 0.725 |
| Hypertrophic scarring | 3.1% (25) | 1.6% (7) | 4.7% (18) | 0.022 |
| Scar revision | 2.5% (20) | 1.4% (6) | 3.6% (14) | 0.07 |
| Fat necrosis | 2.3% (19) | 2.8% (12) | 1.8% (7) | 0.476 |
| Complication by patient | ||||
| Number of patients | 406 | 213 | 193 | |
| Readmission | 2.5% (10) | 1.9% (4) | 3.1% (6) | 0.632 |
| Postoperative DVT/PE | 0.5% (2) | 0.0% (0) | 1.0% (2) | 0.436 |
Frequencies of categorical variables are reported as % (n).
Continuous variable (PODs to last follow-up) is reported as mean ± standard deviation.
Boldface indicates statistical significance (p<0.05).
Only two patients had a postoperative DVT/PE, both of whom did not receive TXA. These events occurred over 4 months after each breast reduction. Additionally, both of these patients had significant comorbidities for clotting, including a current diagnosis of breast cancer and age over 70 years. The patient with a DVT did not require inpatient admission and was treated with oral apixaban. The patient with a PE was admitted for 1 day for treatment with IV heparin, then bridged to oral apixaban.
In the univariate analysis, IV TXA use alone significantly predicted decreased rates of dehiscence (OR = 0.30; 95% CI 0.18–0.51; p < 0.001), hypertrophic scarring (OR = 0.35; 95% CI 0.15–0.84; p = 0.013), and scar revision (OR = 0.40; 95% CI 0.16–1.0; p = 0.044) (Table 4). Considering the baseline discrepancies in demographic and pre- or peri‑operative characteristics between the TXA and non-TXA groups, additional regression models were created to account for these differences and better clarify the relationship between TXA use and wound complications. Once demographic characteristics were adjusted for, IV TXA use was still a significant predictor of lower dehiscence rates (aOR = 0.32; 95% CI 0.19–0.54; p < 0.001), hypertrophic scarring (aOR = 0.33; 95% CI 0.14–0.77; p = 0.009) and scar revision (aOR = 0.32; 95% CI 0.13–0.80; p = 0.014). When adjusting for operative characteristics, IV TXA was a significant predictor of decreased rates of dehiscence (aOR = 0.45; 95% CI 0.25–0.79; p = 0.005) and scar revision (aOR = 0.29; 95% CI 0.11–0.76; p = 0.013) but was no longer a significant predictor for hypertrophic scarring (p = 0.160). IV TXA use continued to be a significant predictor of decreased rates of these complications after adjusting for both operative and demographic variables, with adjusted odds ratios of 0.50 (95% CI 0.28–0.91; p = 0.023) for dehiscence and 0.30 (95% CI 0.12–0.76; p = 0.015) for scar revision.
Table 4.
Multivariable regression assessing association of TXA use and individual wound complications.a
| Unadjusted OR | p-value | Model 1 aOR | p-value | Model 2 aOR | p-value | Model 3 aOR | p-value | |
|---|---|---|---|---|---|---|---|---|
| Seroma | 0.37 (0.10–1.28) | 0.098 | 0.54 (0.16–1.80) | 0.344 | 0.63 (0.18–2.25) | 0.496 | 1.01 (0.28–3.63) | 0.985 |
| Hematoma | 0.43 (0.18–1.04) | 0.052 | 0.44 (0.19–1.04) | 0.064 | 0.49 (0.19–1.27) | 0.150 | 0.54 (0.19–1.56) | 0.257 |
| Dehiscence | 0.30 (0.18–0.51) | <0.001 | 0.32 (0.19–0.54) | <0.001 | 0.45 (0.25–0.79) | 0.005 | 0.50 (0.28–0.91) | 0.023 |
| Infection | 1.50 (0.46–4.87) | 0.492 | 1.61 (0.53–4.84) | 0.433 | 1.53 (0.46–5.06) | 0.507 | 1.52 (0.48–4.78) | 0.528 |
| Nipple insensitivity | 1.15 (0.65–2.07) | 0.624 | 1.07 (0.60–1.91) | 0.820 | 0.72 (0.38–1.36) | 0.323 | 0.68 (0.36–1.26) | 0.236 |
| Hypertrophic scarring/keloid | 0.35 (0.15–0.84) | 0.013 | 0.33 (0.14–0.77) | 0.009 | 0.51 (0.20–1.29) | 0.160 | 0.50 (0.21–1.19) | 0.132 |
| Scar revision | 0.40 (0.16–1.0) | 0.044 | 0.32 (0.13–0.80) | 0.014 | 0.29 (0.11–0.76) | 0.013 | 0.30 (0.12–0.76) | 0.015 |
| Fat necrosis | 1.53 (0.62–3.81) | 0.359 | 1.41 (0.58–3.45) | 0.473 | 1.35 (0.51–3.61) | 0.560 | 1.30 (0.50–3.39) | 0.622 |
OR, odds ratio; aOR, adjusted odds ratio.
OR and aOR are reported with 95% confidence interval in parentheses.
Model 1 adjusts for demographic characteristics: age, smoking history, Caprini Score, preoperative BMI, and history of weight loss.
Model 2 adjusts for operative characteristics: weight of tissue removed, procedure duration, pedicle choice, drain usage.
Model 3 adjusts for all the factors that were included in both Models 1 and 2.
Includes re-classified Smoking History (never vs. current/former) and pedicle choice (superomedial vs. not superomedial).
Boldface indicates statistical significance (p<0.05).
Firth’s bias correction was used to accommodate the low frequencies of complications.
As the frequency of barbed suture use differed between TXA and non-TXA groups, a second regression analysis was performed with barbed suture included as a covariate in the operative characteristics model (Model 2) and the combined model (Model 3). In this analysis, IV TXA use was only a significant predictor of decreased rates of scar revision for Model 2 (aOR = 0.26; 95% CI 0.09–0.75; p = 0.018) and Model 3 (aOR = 0.27; 95% CI 0.09–0.77; p = 0.026). While IV TXA was not found to be a significant predictor of dehiscence, among the covariates, it was found that barbed suture usage was a significant predictor of decreased rates of dehiscence in both Model 2 (aOR = 0.26; 95% CI 0.10–0.72; p = 0.007) and Model 3 (aOR = 0.33; 95% CI 0.13–0.88; p = 0.023) (Table 5). This discrepancy in significance prompted our additional regression analysis on breasts that had their incisions closed with barbed sutures. In this analysis, IV TXA was a significant predictor of decreased rates of scar revision after all adjustments: demographics (aOR = 0.32; 95% CI 0.11–0.90; p = 0.049), operative characteristics (aOR = 0.22; 95% CI 0.07–0.67; p = 0.012), and a combination of both (aOR = 0.16; 95% CI 0.05–0.52; p = 0.005) (Table 6). In addition, a close association was found between IV TXA usage and decreased hematoma rates, with adjusted odds ratios that approached significance across all models.
Table 5.
Output of logistic regression (A) Model 2 and (B) Model 3 predicting occurrence of dehiscence in breasts closed with and without barbed suture as an independent variable.
| (A) Model 2 | ||||
|---|---|---|---|---|
| Without barbed suture | ||||
| Coefficient | Standard error | aOR (95% CI) | p-value | |
| TXA use | −0.80 | 0.29 | 0.45 (0.25–0.79) | 0.005 |
| Weight removed | 0.08 | 0.08 | 1.08 (0.92–1.27) | 0.409 |
| Procedure duration | 0.30 | 0.11 | 1.36 (1.08–1.70) | 0.009 |
| Pedicle choice | −1.13 | 0.33 | 0.32 (0.17–0.61) | <0.001 |
| Drain usage | 0.38 | 0.27 | 1.46 (0.85–2.49) | 0.173 |
| Intercept | −1.71 | 0.19 | 0.18 (0.13–0.26) | <0.001 |
| With barbed suture | ||||
| TXA use | −0.08 | 0.42 | 0.92 (0.40–2.11) | 0.855 |
| Weight removed | 0.09 | 0.08 | 1.09 (0.93–1.29) | 0.368 |
| Procedure duration | 0.18 | 0.12 | 1.19 (0.94–1.52) | 0.152 |
| Pedicle choice | −0.62 | 0.39 | 0.54 (0.25–1.17) | 0.126 |
| Drain usage | 0.62 | 0.29 | 1.86 (1.05–3.30) | 0.035 |
| Barbed suture usage | −1.33 | 0.51 | 0.26 (0.10–0.72) | 0.007 |
| Intercept | −1.55 | 0.19 | 0.21 (0.15–0.30) | <0.001 |
| (B) Model 3 | ||||
|---|---|---|---|---|
| Without barbed suture | ||||
| Coefficient | Standard error | aOR (95% CI) | p-value | |
| TXA use | −0.69 | 0.30 | 0.50 (0.28–0.91) | 0.023 |
| Age | −0.12 | 0.14 | 0.89 (0.67–1.17) | 0.403 |
| Smoking history | −0.85 | 0.29 | 0.43 (0.24–0.75) | 0.004 |
| Preoperative BMI | 0.06 | 0.03 | 1.06 (1.0 – 1.12) | 0.045 |
| Caprini score | 0.33 | 0.10 | 1.39 (1.15–1.67) | 0.002 |
| History of weight loss | −0.42 | 0.37 | 0.66 (0.32–1.37) | 0.256 |
| Weight removed | 0.02 | 0.09 | 1.02 (0.85–1.21) | 0.860 |
| Procedure duration | 0.32 | 0.11 | 1.38 (1.10–1.72) | 0.006 |
| Pedicle choice | −1.07 | 0.33 | 0.34 (0.18–0.66) | 0.001 |
| Drain usage | 0.03 | 0.29 | 1.03 (0.59–1.81) | 0.916 |
| Intercept | −4.17 | 1.06 | 0.02 (0.001–0.12) | <0.001 |
| With barbed suture | ||||
| TXA use | −0.11 | 0.41 | 0.89 (0.40–2.01) | 0.791 |
| Age | −0.12 | 0.14 | 0.89 (0.67–1.18) | 0.414 |
| Smoking history | −0.85 | 0.29 | 0.43 (0.24–0.75) | 0.004 |
| Preoperative BMI | 0.05 | 0.03 | 1.05 (0.99–1.12) | 0.078 |
| Caprini score | 0.30 | 0.10 | 1.36 (1.12–1.64) | 0.003 |
| History of weight loss | −0.49 | 0.38 | 0.61 (0.29–1.29) | 0.190 |
| Weight removed | 0.04 | 0.09 | 1.04 (0.87–1.23) | 0.713 |
| Procedure duration | 0.20 | 0.12 | 1.22 (0.96–1.56) | 0.106 |
| Pedicle choice | −0.70 | 0.39 | 0.50 (0.23–1.06) | 0.079 |
| Drain usage | 0.28 | 0.31 | 1.32 (0.72–2.41) | 0.374 |
| Barbed suture usage | −1.10 | 0.50 | 0.33 (0.13–0.88) | 0.023 |
| Intercept | −3.70 | 1.07 | 0.02 (0.003–0.20) | <0.001 |
Boldface indicates statistical significance (p<0.05).
Table 6.
Multivariable regression assessing association of TXA use and individual wound complications among patients who had their incisions closed with barbed suture.a
| Unadjusted OR | p-value | Model 1 aOR | p-value | Model 2 aOR | p-value | Model 3 aOR | p-value | |
|---|---|---|---|---|---|---|---|---|
| Seroma | 1.71 (0.08–35.8) | 0.714 | 1.85 (0.20–17.0) | 0.664 | 1.75 (0.15–21.5) | 0.694 | 1.56 (0.20–12.3) | 0.768 |
| Hematoma | 0.38 (0.13–1.11) | 0.086 | 0.40 (0.15–1.08) | 0.10 | 0.31 (0.11–0.89) | 0.041 | 0.34 (0.12–0.92) | 0.057 |
| Dehiscence | 1.58 (0.56–4.47) | 0.365 | 1.44 (0.53–3.94) | 0.483 | 1.28 (0.46–3.62) | 0.637 | 1.28 (0.46–3.51) | 0.651 |
| Infection | 1.51 (0.25–8.99) | 0.639 | 1.32 (0.28–6.36) | 0.753 | 2.22 (0.39–12.7) | 0.383 | 2.13 (0.41–11.0) | 0.429 |
| Nipple Insensitivity | 0.68 (0.34–1.35) | 0.278 | 0.64 (0.33–1.27) | 0.223 | 0.81 (0.40–1.62) | 0.564 | 0.78 (0.39–1.54) | 0.487 |
| Hypertrophic Scarring/Keloid | 1.72 (0.29–10.0) | 0.524 | 1.54 (0.33–7.30) | 0.614 | 1.34 (0.26–6.77) | 0.737 | 1.31 (0.28–5.98) | 0.762 |
| Scar Revision | 0.33 (0.11–1.0) | 0.055 | 0.32 (0.11–0.90) | 0.049 | 0.22 (0.07–0.67) | 0.012 | 0.16 (0.05–0.52) | 0.005 |
| Fat Necrosis | 1.23 (0.37–4.07) | 0.735 | 0.88 (0.28–2.79) | 0.840 | 0.69 (0.20–2.40) | 0.582 | 0.51 (0.15–1.70) | 0.337 |
OR, odds ratio; aOR, adjusted odds ratio.
OR and aOR are reported with 95% confidence interval in parentheses.
Model 1 adjusts for demographic characteristics: age, smoking history, Caprini Score, preoperative BMI, and history of weight loss.
Model 2 adjusts for operative characteristics: weight of tissue removed, procedure duration, pedicle choice, drain usage.
Model 3 adjusts for all the factors that were included in both Models 1 and 2.
Includes re-classified Smoking History (never vs. current/former) and pedicle choice (superomedial vs. not superomedial).
Boldface indicates statistical significance (p<0.05).
Firth’s bias correction was used to accommodate the low frequencies of complications.
Discussion
This is the largest study to date that analyzes the use of IV TXA in reduction mammoplasty. We found that IV TXA significantly reduces the rates of wound dehiscence and scar revision after accounting for multiple pre- and peri‑operative factors. However, we also found that these findings are no longer significant when factoring in wound-saving measures, like closure with barbed suture. Therefore, we conclude that IV TXA only has a beneficial effect on postoperative outcomes after reduction mammoplasty when barbed sutures are not used for closure. Importantly, our entire patient cohort had no significant increase in thrombotic events despite the use of TXA.
TXA (trans-4-aminomethylcyclohexane-1-carboxylic acid) is an antifibrinolytic agent that has gained increasing popularity in high-risk procedures involving blood loss. Its mechanism involves competitively inhibiting plasminogen activation to plasmin and preventing the enzymatic breakdown of the fibrin clot, ultimately improving clot stability.2,10 It also has anti-inflammatory properties and can reduce vascular permeability, giving it the theoretical benefit of preventing infection.6 Currently, IV TXA is only FDA-approved for hemophilia and to prevent hemorrhage following tooth extraction. Oral TXA is only FDA-approved for cyclic heavy menstrual bleeding.11 Possible side effects associated with TXA include thrombosis when used concomitantly with other anticoagulants, hypersensitivity reactions, or headaches. Considering its overall low risk of adverse events and zero contraindications except for previous subarachnoid hemorrhage or hypersensitivity, the use of intraoperative TXA is expanding far beyond its original indications.
TXA can come in three forms: oral, IV, and topical. Although oral administration is low-cost, it is rarely used intraoperatively due to its prominent logistical challenge of increased absorption time.12 Previous randomized controlled trials have noted that IV and topical TXA confer similar efficacy.7,8,12 Physicians then tend to decide between these routes based on safety and practical considerations. Concerns about systemic absorption with IV TXA have led some clinicians to prefer topical administration in patients at higher risk for thromboembolic events. In contrast, IV TXA could be selected for its ease of administration and consistent systemic levels.13 Therefore, it is up to the discretion of the senior surgeon to decide the administration method of TXA. Regardless of the administration method, the literature on TXA use is extensive, particularly in specialties such as orthopedics and neurosurgery. In total knee, shoulder, hip, and ankle arthroplasties, IV and topical TXA use has been shown to reduce blood volume loss, changes in hemoglobin and hematocrit, drain output, and the incidence of joint infection.3,14, 15, 16, 17, 18 Neurosurgeons have also explored the use of TXA, particularly in spine surgeries, showing that among those undergoing open spine surgery, intraoperative TXA significantly reduces total blood loss, perioperative blood loss, postoperative drainage, hospital stay, and total blood transfusion volume.4 As such, it is evident that TXA has vast potential to confer lower peri‑ and postoperative complications, reducing operative time and improving patient outcomes.
For plastic surgery procedures, TXA has been primarily supported for its use in craniofacial and cosmetic facial surgeries. A large meta-analysis conducted by Brown et al. summarizes the current standard of practice for TXA in plastic surgery cases, showing a “significant association between TXA administration, reduced hematoma rates, and blood loss” in rhinoplasty, facelifts, and blepharoplasty.2,5 It has also been demonstrated that the use of IV TXA in rhinoplasty decreases the rates of ecchymosis and eyelid edema.19 Similarly, ample academic evidence from craniofacial procedures, such as fronto-orbital advancement, cranial vault remodeling, and cleft palate surgery, demonstrates decreased blood loss rates with the use of IV TXA.2,20, 21, 22 These beg the question of why there is not as much evidence-based support for TXA use in breast reduction, one of the most common operations in plastic surgery.
The current research surrounding TXA in breast surgery primarily emphasizes its benefits in breast reconstruction and mastectomy. Meta-analyses indicate that administering perioperative systemic IV TXA in patients undergoing mastectomy, with or without immediate implant or free-flap-based breast reconstruction, significantly reduces the risk of hematoma and seroma formation.23, 24, 25 Another study by Safran et al., which specifically examined the use of topical TXA in bilateral mastectomies with immediate direct-to-implant reconstruction, found that soaking the mastectomy bed with 3% topical TXA before implant insertion reduces drain output and operative hematomas.26 A meta-analysis by Buheiri et al. expanded the existing literature investigating topical and IV TXA, and provided further support that either method of administration reduces hematoma formation. However, only a minority of the studies included focused on reduction mammaplasty with the remaining majority reporting on oncogenic breast surgeries and reconstructions.27
Thus, there remained an opportunity to deepen the understanding of the utility of TXA in reduction mammoplasty, especially since existing findings substantiate conflicting conclusions. A recent review by Pontes et al. on the impact of TXA on reduction mammoplasty outcomes only found seven studies that analyzed this specific cohort.6 They found that consistently across all studies, including a total of 2232 breasts, there were no adverse events associated with TXA, such as DVT/PE, seizures, or hypersensitivity reactions. Some studies analyzing topical TXA use found no association with fewer complications or lower drain output.28,29 Meanwhile, others report reduced drain fluid production or a significantly decreased number of postoperative hematomas requiring evacuation.30, 31, 32 In a study by Om et al., which analyzed IV TXA use in a small sample of patients undergoing bilateral mastectomy, the authors found that it was associated with significantly lower rates of hematoma; however, it did not affect intraoperative blood loss or operative time.33 Conversely, a retrospective review of IV administration of TXA in reduction mammoplasty did not identify a decrease in postoperative hematoma formation.34 These inconsistencies among groups with small sample sizes in the current literature reinforce the need for research with larger patient cohorts when evaluating the utility of TXA for reduction mammoplasty.
Ultimately, our study found that systemic IV TXA use confers statistically significantly lower rates of wound dehiscence and scar revision, provided that conventional suture closure is used. We found that the use of barbed suture for closure after the reduction mammoplasty confounds the improvement in postoperative healing. These conclusions account for differences in preoperative characteristics, like age, smoking history, Caprini Score, preoperative BMI, and weight loss history, as well as operative and postoperative characteristics, including the tissue weight removed, procedure duration, pedicle choice, and use of drains. Supported by the largest cohort who specifically received IV TXA, these findings are a vital addition to the growing literature on factors that impact outcomes following reduction mammoplasty. We hypothesize that our findings are due to TXA’s ability to minimize subclinical hematomas and seromas that do not require evacuation or readmission but lead to local wound space expansion, dehiscence, and, ultimately, scar revision. We also believe this association may be due to TXA’s anti-inflammatory properties, which reduce edema and local wound dehiscence that would have led to scarring.6 Also, aligning with the previous literature, we found no association with TXA-affiliated postoperative complications like DVT, PE, or hypersensitivity.
As mentioned earlier, when considering operative choices intentionally made to prevent dehiscence and scarring, such as using barbed sutures for closure, TXA no longer significantly impacts wound dehiscence. This leads us to conclude that while TXA may be moderately effective in minimizing the formation of small fluid collections, the closure technique is more critical to prevent adverse outcomes.
Our study does include some inherent limitations. One is our inclusion of six senior surgeons whose unique surgical training and preferences guided decisions regarding the use of drains, barbed sutures, and timing of TXA administration in the absence of a standardized protocol. Although this could have been confounding, we completed a detailed multivariable regression that accounts for differences in technique, pedicle choice, and skin closure that might have otherwise skewed the results. The reasoning for surgeons’ decisions and preferences was not further explored in this study, but it may be an avenue for future study. Another limitation of retrospective studies is that we were unable to adjust our sample demographics; however, we mitigated this by using regression analysis. Ultimately, we adjusted for as many confounding variables as possible in our dataset. This study was limited in its analysis of TXA’s effects on drain output and, consequently, on drain duration, as noted in prior literature. This is because the drain output quantity was inconsistently reported or frequently unavailable in the patients’ records. Given the potential impact of TXA on drain output, as suggested by prior studies, any future prospective studies on the effect of TXA on breast surgery outcomes should include this measure. Finally, our study was conducted at only one academic institution, making its generalizability more challenging. Despite these limitations, this study provides evidence for the use of IV TXA in reduction mammoplasty to minimize postoperative complications and improve cosmesis when using conventional closure techniques. Future studies are necessary to broaden the scope of this single-institutional analysis.
Conclusion
This study, the largest to date on the use of IV TXA during reduction mammoplasty, found that TXA use is predictive of reduced rates of wound dehiscence and the need for scar revision after controlling for demographic and perioperative factors. However, using wound closure techniques like barbed sutures negates this benefit. Although TXA has the potential to decrease postoperative complications after reduction mammoplasty, its impact is not significant enough to outweigh the importance of wound closure choice.
Ethical approval
Brown University Health Institutional Review Board (IRB), Providence, RI. IRB Net ID #1910469-4.
Declaration of competing interests
The authors have no financial conflicts of interest to disclose.
Funding
The authors have no sources of funding to disclose.
Footnotes
Presented at: AAPS Annual Meeting, Podium Presentation, May 24–27th, 2025, Austin, TX.
References
- 1.Morrison K.A., Frey J.D., Karp N., Choi M. Revisiting reduction mammaplasty: complications of oncoplastic and symptomatic macromastia reductions. Plast Reconstr Surg. 2023;151(2):267–276. doi: 10.1097/PRS.0000000000009828. [DOI] [PubMed] [Google Scholar]
- 2.Brown S., Brown T., Rohrich RJ. Clinical applications of tranexamic acid in plastic and reconstructive surgery. Plast Reconstr Surg. 2024;154(6):1253e–1263e. doi: 10.1097/PRS.0000000000011288. [DOI] [PubMed] [Google Scholar]
- 3.Laungani D., Porto J.R., Haase L., Smith K., Chen R., Gillespie R. Tranexamic acid in total shoulder arthroplasty: a scoping review of current practices and future directions. JBJS Rev. 2024;12(6) doi: 10.2106/JBJS.RVW.24.00035. [DOI] [PubMed] [Google Scholar]
- 4.Liu Z.G., Yang F., Zhu Y.H., Liu G.C., Zhu Q.S., Zhang BY. Is tranexamic acid beneficial in open spine surgery? And its effects vary by dosage, age, sites, and locations: a meta-analysis of randomized controlled trials. World Neurosurg. 2022;166:141–152. doi: 10.1016/j.wneu.2022.07.044. [DOI] [PubMed] [Google Scholar]
- 5.Brown S., Yao A., Taub PJ. Antifibrinolytic agents in plastic surgery: current practices and future directions. Plast Reconstr Surg. 2018;141(6):937e–949e. doi: 10.1097/PRS.0000000000004421. [DOI] [PubMed] [Google Scholar]
- 6.Pontes A., Barreiro D., Costa-Ferreira A. The impact of tranexamic acid administration in reduction mammaplasty: a systematic review. Ann Plast Surg. 2025;94(3):370–377. doi: 10.1097/SAP.0000000000004184. [DOI] [PubMed] [Google Scholar]
- 7.Abdel M.P., Chalmers B.P., Taunton M.J., et al. Intravenous versus topical tranexamic acid in total knee arthroplasty: both effective in a randomized clinical trial of 640 patients. J Bone Joint Surg Am. 2018;100(12):1023–1029. doi: 10.2106/JBJS.17.00908. [DOI] [PubMed] [Google Scholar]
- 8.Hsu Y.C., Hsu A.H.S., Wu C.T., Tan T.L., Wang J.W., Kuo FC. Association between IV and topical tranexamic acid use and periprosthetic joint infections in hip and knee arthroplasty: a retrospective study. BMC Musculoskelet Disord. 2024;25(1):1033. doi: 10.1186/s12891-024-08080-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Francalancia S., Lou M., McIntire D., et al. Evaluation of permanent nipple-areolar complex sensitivity loss following reduction mammaplasty. Aesthet Surg J. 2025;45(2):NP50–NP56. doi: 10.1093/asj/sjae215. [DOI] [PubMed] [Google Scholar]
- 10.Chauncey J.M., Wieters JS. StatPearls. StatPearls Publishing; 2025. Tranexamic acid. [PubMed] [Google Scholar]
- 11.U.S. Food and Drug Administration. Tranexamic acid: highlights of prescribing information. https://www.accessdata.fda.gov/drugsatfda_docs/label/2021/019281s047lbl.pdf. Accessed February 15, 2025.
- 12.Hootsmans N.A.M., Vellanky S., Grady-Benson J., Cremins MS. A randomized trial of three routes of tranexamic acid administration in total knee arthroplasty. Orthopedics. 2023;46(5):285–290. doi: 10.3928/01477447-20230224-04. [DOI] [PubMed] [Google Scholar]
- 13.Xie J., Hu Q., Huang Q., Ma J., Lei Y., Pei F. Comparison of intravenous versus topical tranexamic acid in primary total hip and knee arthroplasty: an updated meta-analysis. Thromb Res. 2017;153:28–36. doi: 10.1016/j.thromres.2017.03.009. [DOI] [PubMed] [Google Scholar]
- 14.Sun C., Zhang X., Ma Q., Tu Y., Cai X., Zhou Y. Impact of tourniquet during total knee arthroplasty when tranexamic acid was used: a meta-analysis of randomized controlled trials. J Orthop Surg Res. 2022;17(1):18. doi: 10.1186/s13018-021-02898-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Abildgaard J.T., McLemore R., Hattrup SJ. Tranexamic acid decreases blood loss in total shoulder arthroplasty and reverse total shoulder arthroplasty. J Shoulder Elbow Surg. 2016;25(10):1643–1648. doi: 10.1016/j.jse.2016.02.002. [DOI] [PubMed] [Google Scholar]
- 16.Zhong J., Chen G., Si Y., Xie J. Tranexamic acid administration in total shoulder arthroplasty: a systematic review and meta-analysis. BMC Musculoskelet Disord. 2024;25(1):1042. doi: 10.1186/s12891-024-08159-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Ghobrial P.G., Eikani C.K., Schmitt D.R., Brown N.M., Pinzur M.S., Schiff AP. Safety and efficacy of tranexamic acid in total ankle arthroplasty. Foot Ankle Spec. 2023 doi: 10.1177/19386400231207276. [DOI] [PubMed] [Google Scholar]
- 18.Elmenawi K.A., Mohamed F.A.E., Poilvache H., Prokop L.J., Abdel M.P., Bedard NA. Association between tranexamic acid and decreased periprosthetic joint infection risk in patients undergoing total hip and knee arthroplasty: a systematic review and meta-analysis of over 2 million patients. J Arthroplasty. 2024;39(9):2389–2394.e2. doi: 10.1016/j.arth.2024.04.033. [DOI] [PubMed] [Google Scholar]
- 19.McGuire C., Nurmsoo S., Samargandi O.A., Bezuhly M. Role of tranexamic acid in reducing intraoperative blood loss and postoperative edema and ecchymosis in primary elective rhinoplasty: a systematic review and meta-analysis. JAMA Facial Plast Surg. 2019;21(3):191–198. doi: 10.1001/jamafacial.2018.1737. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Crantford J.C., Wood B.C., Claiborne J.R., et al. Evaluating the safety and efficacy of tranexamic acid administration in pediatric cranial vault reconstruction. J Craniofac Surg. 2015;26(1):104–107. doi: 10.1097/SCS.0000000000001271. [DOI] [PubMed] [Google Scholar]
- 21.Engel M., Bodem J.P., Busch C.J., et al. The value of tranexamic acid during fronto-orbital advancement in isolated metopic craniosynostosis. J Craniomaxillofac Surg. 2015;43(7):1239–1243. doi: 10.1016/j.jcms.2015.05.004. [DOI] [PubMed] [Google Scholar]
- 22.Durga P., Raavula P., Gurajala I., et al. Evaluation of the efficacy of tranexamic acid on the surgical field in primary cleft palate surgery on children-A prospective, randomized clinical study. Cleft Palate Craniofac J. 2015;52(5):e183–e187. doi: 10.1597/14-263. [DOI] [PubMed] [Google Scholar]
- 23.Liechti R., van de Wall B.J.M., Hug U., Fritsche E., Franchi A. Tranexamic acid use in breast surgery: a systematic review and meta-analysis. Plast Reconstr Surg. 2023;151(5):949–957. doi: 10.1097/PRS.0000000000010071. [DOI] [PubMed] [Google Scholar]
- 24.Huynh M.N.Q., Wong C.R., McRae M.C., Voineskos S., McRae MH. The effects of tranexamic acid in breast surgery: a systematic review and meta-analysis. Plast Reconstr Surg. 2023;152(6):993e–1004e. doi: 10.1097/PRS.0000000000010479. [DOI] [PubMed] [Google Scholar]
- 25.Weissler J.M., Banuelos J., Jacobson S.R., et al. Intravenous tranexamic acid in implant-based breast reconstruction safely reduces hematoma without thromboembolic events. Plast Reconstr Surg. 2020;146(2):238–245. doi: 10.1097/PRS.0000000000006967. [DOI] [PubMed] [Google Scholar]
- 26.Safran T., Vorstenbosch J., Viezel-Mathieu A., Davison P., Dionisopoulos T. Topical tranexamic acid in breast reconstruction: a double-blind randomized controlled trial. Plast Reconstr Surg. 2023;152(4):699–706. doi: 10.1097/PRS.0000000000010322. [DOI] [PubMed] [Google Scholar]
- 27.Buheiri A.R., Tveskov L., Dines L.M., Bagge J.D., Möller S., Bille C. Tranexamic Acid in breast surgery—a systematic review and meta-analysis. Clin Breast Cancer. 2025;25(5):e496–e510. doi: 10.1016/j.clbc.2025.01.011. [DOI] [PubMed] [Google Scholar]
- 28.Weissler J.M., Kuruoglu D., Antezana L., et al. Efficacy of tranexamic acid in reducing seroma and hematoma formation following reduction mammaplasty. Aesthet Surg J. 2022;42(6):616–625. doi: 10.1093/asj/sjab399. [DOI] [PubMed] [Google Scholar]
- 29.Yao A., Wang F., Benacquista T., et al. Topical tranexamic acid does not reduce hematoma in reduction mammaplasty: a double-blind randomized controlled trial. Plast Reconstr Surg. 2024;154(1):30–37. doi: 10.1097/PRS.0000000000010952. [DOI] [PubMed] [Google Scholar]
- 30.Ausen K., Fossmark R., Spigset O., Pleym H. Randomized clinical trial of topical tranexamic acid after reduction mammoplasty. Br J Surg. 2015;102(11):1348–1353. doi: 10.1002/bjs.9878. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Sipos K., Kämäräinen S., Kauhanen S. Topical tranexamic acid reduces postoperative hematomas in reduction mammaplasties. J Plast Reconstr Aesthet Surg. 2023;83:172–179. doi: 10.1016/j.bjps.2023.04.039. [DOI] [PubMed] [Google Scholar]
- 32.Nichols G., Andrade E., Gregoski M., Herrera F.A., Armstrong MB. Local infiltration of tranexamic acid in breast reduction for symptomatic macromastia. Ann Plast Surg. 2023;90(6S Suppl 4):S366–S370. doi: 10.1097/SAP.0000000000003456. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Om A., Marxen T., Kebede S., Losken A. The usage of intravenous tranexamic acid in reduction mammaplasty safely reduces hematoma rates. Ann Plast Surg. 2023;90(6S Suppl 4):S371–S374. doi: 10.1097/SAP.0000000000003296. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Habarth-Morales T.E., Isch E., Zavitsanos A.P., et al. The use of tranexamic acid in breast reduction and abdominoplasty: a review of a multicenter federated electronic health record database. Aesthet Surg J Open Forum. 2024;6:ojae077. doi: 10.1093/asjof/ojae077. [DOI] [PMC free article] [PubMed] [Google Scholar]