Bringing doctors together

or

A free, private network for verified physicians

Leading voices

Dr. Jad M Abdelsattar, M.D, FACS

Breast Surgical Onc · University of Oxford

View

Dr. Lauren Kopicky

Breast Surgical Onc · Cleveland Clinic

View

Dr. Jennifer Plichta

Breast Surgical Onc · Duke University

View

Dr. Sarah Blair

Breast Surgical Onc · UCSD

View

Dr. Karen Goodwin

Breast Surgical Onc · UCSF

View

Dr. Christin Collins

Breast Surgical Onc · MetroHealth

View

Dr. Philip Schauer

Bariatric Surgery · PBRC

View

Dr. Todd Rosengart

Cardiac Surgery · BCM

View

Dr. Larry Kaiser

Thoracic Surgery · UPenn

View

Dr. Neri Cohen

Thoracic Surgery · XVIVO

View
  • Discover your community

  • Stay current

  • A profile that represents you

  • Find your people

About RoonPrivacyTerms

    Overview

    Long-Term Outcomes of Lymphedema After Immediate Lymphatic Reconstruction Following Axillary Lymph Node Dissection.

    Apr 16, 2025DOI: 10.1245/s10434-025-17301-0Journal ArticleASO

    Written by: Abbas M HassanJohn P HajjJohn P LewisShahnur AhmedCarla S FisherKandice K LudwigRachel M DanforthR Jason VonDerHaarRavinder BambaMary E LesterAladdin H Hassanein

    Find these authors on Roon

    CF
    Dr. Carla Fisher
    Breast Surgical Onc · IUSM

    Patients and Methods

    This was a retrospective cohort study of consecutive adult (> 18 years) female patients with breast cancer who underwent ILR following ALND with/without breast reconstruction at six hospitals within the Indiana University network from 2017 to 2024. Patients were excluded if they underwent sentinel lymph node biopsy (SLNB) without ALND, those with preexisting lymphedema, or secondary causes of swelling, such as deep venous thrombosis or chronic venous insufficiency. This study was approved by Indiana University Institutional Review Board.The following patient-level information was abstracted from the medical records and compared between the study groups: patient demographics and characteristics (age, body mass index [BMI], race, tobacco use, American Society of Anesthesiologists [ASA] classification, cancer type), comorbidities (diabetes mellitus, hypertension, hyperlipidaemia, chronic obstructive pulmonary disease), adjuvant therapy (preoperative and postoperative chemotherapy and/or radiotherapy, radiation location), radiation location (axillary, chest wall/breast, supraclavicular, and/or internal mammary radiation), nodal staging (number of lymph nodes removed and positive lymph nodes), surgical characteristics (type of breast surgery and timing of reconstruction procedures), and intraoperative techniques (number of lymphovenous bypasses performed). Tobacco use was defined as use of any tobacco product within 8 weeks of ALND. Obesity was defined as a BMI > 30 kg/m2.The primary outcome was the prevalence of BCRL following ILR, defined as a difference of ≥ 2 cm between the affected and contralateral limb at two or more contiguous measurement points.37 Lymphedema prevalence in our cohort was compared with previously published pooled estimates of lymphedema after ALND without ILR at time points of less than 12 months, 12 to 24 months, and beyond 24 months.5 Patients received preoperative and postoperative physical or occupational therapy consultations. Preoperative consultations included baseline limb measurements and education on recognizing signs and symptoms of lymphedema. Postoperatively, patients had regular follow-up to monitor for lymphedema and assess limb measurements. Secondary outcomes included identifying predictors of BCRL, assessing postoperative surgical complications, recording instances of extremity cellulitis, and noting patients who used compression garments without meeting the diagnostic criteria for BCRL.Immediate lymphatic reconstruction was performed by five microsurgeons following a standardized surgical approach as previously described.26,38,39 In brief, disrupted lymphatic vessels were visualized using isosulfan blue or fluorescein. Anastomoses were executed using the sleeve technique. A vein graft from a branch of the saphenous was utilized if required to connect the lymphatics to suitable vein branch, typically from the axillary or thoracodorsal.39 Each procedure involved placing a single drain in the axilla and another drain in the breast if reconstruction was performed. Axillary lymph node dissection was performed by eight breast surgeons according to standard oncologic protocols, either through the mastectomy incision or via a separate axillary approach.40
    Categorical variables were presented as frequencies and percentages, with comparisons made using the chi-squared test or Fisher’s exact test, as appropriate. Continuous variables were reported as means with standard deviations or medians with interquartile ranges and analyzed by using the t-test for normally distributed data or the Mann-Whitney U test for nonparametric data. A one-sample proportion Z-test was employed to compare lymphedema prevalence in our cohort with previously published pooled estimates following ALND without ILR. Univariate and multivariable regression analyses were conducted to calculate odds ratios (ORs) and 95% confidence intervals (CIs) to identify risk factors present at the time of surgery. Parsimonious multivariable models were developed by using stepwise selection based on the lowest Akaike information criterion. All statistical tests were two-sided, and a p-value < 0.05 was considered statistically significant. Data analyses were performed using SAS Enterprise Guide, version 9.4 (SAS Institute Inc., Cary, NC).

    Results

    We identified 185 consecutive patients who underwent ILR following ALND during the study period. Of those, nine were deceased or lost to follow-up and three had lymphedema prior to ILR. The final analytic sample included 172 patients with mean age 50.9 ± 11.6 years, BMI of 29.5 ± 6.9 kg/m2, and follow-up of 23.1 ± 15.2 months. Among these patients, 12 patients (7.0%) developed lymphedema and 160 (93.0%) did not. When comparing the prevalence of lymphedema in our cohort to previously published pooled estimates of 10,774 patients who underwent ALND without ILR,5 we observed significantly lower rates at all timepoints. Specifically, the prevalence of lymphedema in our ILR cohort was 2.5% at less than 12 months postoperatively, 3.7% between 12 and 24 months, and 7.0% beyond 24 months. In contrast, reported rates for patients undergoing ALND without ILR were 16.5%, 24.6%, and 23.6%, respectively (all p < 0.05) (Fig. 1).Fig. 1Comparison of lymphedema prevalence over time in patients who had ALND with ILR with pooled estimates from ALND alone by Bakri et al. (2023)
    Comparison of lymphedema prevalence over time in patients who had ALND with ILR with pooled estimates from ALND alone by Bakri et al. (2023)
    The majority of patients were White or Caucasian (n= 134; 77.9%), classified as ASA class 3 (n = 95; 55.2%), and diagnosed with infiltrating ductal carcinoma (n = 127; 73.8%). Table 1 summarizes the baseline characteristics of patients who did and did not develop lymphedema following ILR. Patients who developed lymphedema following ILR had significantly longer median (range) follow-up time (30.5 [range 9.0–76.0] vs. 19.0 [12.0–75.0] months, p= 0.018) compared with those who did not. Black/African American patients had significantly higher lymphedema rates compared with White patients (18.8% vs. 4.5%, p= 0.005). There were no other significant differences between the two groups regarding demographic factors, preexisting comorbidities, or receipt of adjuvant therapies.Table 1Patient demographicsVariableAll ptsStudy groups*pNo LymphedemaLymphedemaPatients, n (%)172160 (93.0)12 (7.0)Age, years, mean ± SD50.9 ± 11.651.2 ± 11.647.5 ± 11.70.315Body mass index, mean ± SD29.5 ± 6.929.4 ± 7.031.6 ± 5.70.230Obesity, n (%)72 (41.9)66 (41.3)6 (50.0)0.553Race, n (%)0.005White or Caucasian134 (77.9)128 (95.5)6 (4.5)Black or African American32 (18.6)26 (81.3)6 (18.8)Asian6 (3.5)6 (100.0)0 (0.0)ASA classification, n (%)0.823277 (44.7)72 (45.0)5 (41.7)395 (55.2)88 (55.0)7 (58.3)Length of follow-up, monthsmean ± SD23.1 ± 15.222.2 ± 14.434.9 ± 21.10.064Median (range)20.0 (6.0–76.0)19.0 (12.0–75.0)30.5 (9.0–76.0)0.018Tobacco use, n (%)18 (10.5)16 (10.0)2 (16.7)0.467Medical comorbidity, n (%)Hypertension50 (29.1)45 (28.1)5 (41.7)0.319Diabetes mellitus26 (15.1)24 (15.0)2 (16.7)0.876Hyperlipidemia42 (24.4)39 (24.4)3 (25.0)0.961Chronic obstructive pulmonary disease5 (2.9)4 (2.5)1 (8.3)0.563Cancer type, n (%)0.309Infiltrating ductal carcinoma127 (73.8)116 (72.5)11 (91.7)Infiltrating lobular carcinoma18 (10.5)18 (11.3)0 (0.0)Other27 (15.7)26 (16.3)1 (8.3)ASA American Society of Anesthesiologists
    Patient demographics
    ASA American Society of Anesthesiologists
    Most patients underwent mastectomy, ALND with breast reconstruction (n= 97; 57.7%), 26.2% had mastectomy with ALND without reconstruction, and 16.1% had lumpectomy and ALND without reconstruction. The most common (44.8%, n= 77) reconstruction modality was tissue expander. Surgical characteristics of both groups are detailed in Table 2. Among patients who developed lymphedema, 50.0% were obese compared with 41.3% of those who did not develop lymphedema (p= 0.553). Additionally, 91.7% of patients with lymphedema received postoperative radiotherapy compared to 84.8% of those without lymphedema (p= 0.518). Fifty-five percent (n= 92) of patients received neoadjuvant chemotherapy, whereas only 24.3% received adjuvant chemotherapy. The majority (71.8%, n= 102) of patients had axillary, breast/chest wall with supraclavicular and/or internal mammary radiation, whereas only 28.2% had axillary, breast/chest wall radiation alone. The median number of lymph nodes removed during ALND was 15 (IQR, 10.0–21.0), and median number of positive lymph nodes was 2.0 (IQR, 0.0–4.0). Most patients had a single lymphovenous bypass performed (IQR 1, 1–2). No significant differences were found between patients with and without lymphedema for receiving neoadjuvant and adjuvant chemotherapy (p= 0.779 and p= 0.400, respectively), radiation location (p= 0.976), breast reconstruction type (p= 0.399), timing of reconstruction (p= 0.424), number (p= 0.191) and positivity (p= 0.654) of lymph node removed, or number of bypasses performed (p= 0.949).Table 2Surgical characteristicsVariableAll patientsStudy groups*pNo lymphedemaLymphedemaPatients, n (%)172160 (93.0)12 (7.0)Surgery type, n (%)0.450Lumpectomy and ALND27 (16.1)26 (96.3)1 (3.7)Mastectomy and ALND44 (26.2)42 (95.5)2 (4.6)Mastectomy, ALND and breast reconstruction97 (57.7)88 (90.7)9 (9.3)Chemotherapy, n (%)Preoperative92 (54.4)85 (54.1)7 (58.3)0.779Postoperative41 (24.3)37 (23.6)4 (33.3)0.400Postoperative radiotherapy, n (%)145 (85.3)134 (84.8)11 (91.7)0.518Radiation location, n (%)0.945Axillary and breast/chest wall40 (28.2)37 (92.5)3 (7.5)Axillary, breast/chest wall with supraclavicular and/or internal mammary radiation102 (71.8)94 (92.2)8 (7.8)No. lymph nodes removed during ALND, Median (IQR)15.0 (10.0–21.0)15.9 (10.0–21.0)18.5 (12.0–25.0)0.191No. positive lymph nodes removed during ALND, median (IQR)2.0 (0.0–4.0)2.0 (0.0–4.0)2.0 (0.3–6.5)0.654No. lymphovenous bypasses performed, median (IQR)1 (1–2)1.0 (1.0–2.0)1.0 (1.0–2.0)0.949Breast reconstruction type, n (%)0.399None75 (43.6)72 (45.0)3 (25.0)Prosthetic77 (44.8)70 (43.8)7 (58.3)Autologous20 (11.6)18 (11.3)2 (16.7)Timing of reconstruction, n (%)0.424Immediate13 (13.4)12 (13.6)1 (11.1)Delayed17 (17.5)14 (15.9)3 (33.3)Staged67 (69.1)62 (70.5)5 (55.6)
    Surgical characteristics
    Surgical outcomes for patients with and without lymphedema are presented in Table 3. The rate of breast-related complications was 30.2% with no significant difference between the two groups (p= 0.122). There were no significant differences in tissue expander-related complications (p= 0.161) or donor-site complications (p= 1.000). Fifty-five patients (32.0%) used compression garments at one point during their postoperative course, of whom 43 (78.2%) did not meet the diagnostic criteria for lymphedema. The median time to significant limb swelling was 4.5 (IQR, 1.0–11.3) months. Eight patients (4.7%) developed extremity cellulitis.Table 3Surgical outcomesVariableAll patientsStudy groups*pNo lymphedemaLymphedemaPatients, n (%)172160 (93.0)12 (7.0)Compression use55 (32.0)43 (26.9)12 (100.0)<0.001Time to significant limb swelling, months, Median (IQR)4.5 (1.0–11.3)4.0 (0.0–11.0)1.5 (1.0–6.5)0.678Extremity cellulitis, n (%)8 (4.7)6 (3.8)2 (16.7)0.102Any breast-related complication, n (%)52 (30.2)46 (28.8)6 (50.0)0.122Infection1 (0.6)0 (0.0)1 (8.3)0.067Wound dehiscence8 (4.7)8 (5.0)0 (0.0)0.222Skin flap necrosis10 (5.8)8 (5.0)2 (16.7)0.147Hematoma7 (4.1)6 (3.8)1 (8.3)0.403Seroma2 (1.2)2 (1.3)0 (0.0)1.000Partial flap loss1 (0.6)1 (0.6)0 (0.0)1.000Total flap loss2 (1.2)1 (0.6)1 (8.3)0.135Arterial thrombosis0 (0.0)0 (0.0)0 (0.0)1.000Venous thrombosis1 (0.6)1 (0.6)0 (0.0)1.000Any expander-related complication, n (%)21 (12.2)18 (11.3)3 (25.0)0.161Implant/expander exposure3 (1.7)3 (1.9)0 (0.0)1.000Implant/expander deflation/rupture4 (2.3)3 (1.9)1 (8.3)0.253Capsular contraction4 (2.3)4 (2.5)0 (0.0)1.000Implant/expander explantation15 (8.7)13 (8.1)2 (16.7)0.312Any donor-site complication, n (%)1 (0.6)1 (0.6)0 (0.0)1.000Infection0 (0.0)0 (0.0)0 (0.0)1.000Delayed wound healing1 (0.6)1 (0.6)0 (0.0)1.000Hematoma0 (0.0)0 (0.0)0 (0.0)1.000Seroma0 (0.0)0 (0.0)0 (0.0)1.000IQR interquartile range
    Surgical outcomes
    IQR interquartile range
    Multivariable regression analysis was performed to identify factors associated with the development of lymphedema following ILR (Table 4). Compared with White race, Black/African American race was independently associated with an increased risk of BCRL after ILR (OR, 6.38; p < 0.006).Table 4Univariate and multivariable models for development of lymphedemaUnivariate modelMultivariable modelVariableOR (95% CI)pOR (95% CI)pAge (years)0.97 (0.92–1.02)0.291BMI (kg/m2)1.04 (0.97–1.12)0.296Race, n (%)White or CaucasianRefBlack or African American4.92 (1.48–16.47)0.0106.38 (1.70-23.93)0.006ASA classification2Ref31.15 (0.35–3.76)0.823Tobacco use1.80 (0.36–8.95)0.473Medical comorbidityHypertension1.83 (0.55–6.01)0.325Diabetes mellitus1.13 (0.23–5.50)0.877Hyperlipidemia1.03 (0.26–4.01)0.961Cancer typeOtherRefInfiltrating ductal carcinoma2.47 (0.30–19.94)0.398Breast surgery typeLumpectomy and ALNDRefMastectomy and ALND1.24 (0.11–14.34)0.864Mastectomy, ALND and breast reconstruction2.66 (0.32–21.97)0.364ChemotherapyPreoperative1.19 (0.36–3.90)0.779Postoperative1.62 (0.46–5.70)0.450Postoperative radiotherapy2.41 (0.30–19.41)0.410Radiation locationAxillary and breast/chest wallRefAxillary, breast/chest wall with supraclavicular and/or internal mammary radiation1.02 (0.26–4.07)0.976No. lymph nodes removed during ALND1.03 (0.97–1.10)0.3101.06 (0.99–1.13)0.126No. positive lymph nodes removed during ALND1.03 (0.90–1.20)0.648No. lymphovenous bypasses performed1.14 (0.55–2.39)0.722Postoperative breast complication2.48 (0.76–8.10)0.132Bold values indicate statistically significant findingsOR odds ratio; CI confidence interval; BMI body mass index; ASA American Society of Anesthesiologists
    Univariate and multivariable models for development of lymphedema
    Bold values indicate statistically significant findings
    OR odds ratio; CI confidence interval; BMI body mass index; ASA American Society of Anesthesiologists

    Discussion

    Lymphedema remains a significant problem for many breast cancer patients undergoing ALND, leading to physical discomfort, impaired arm function, psychological distress, and an increased risk of infection.41–43 These challenges underscore the need for preventive strategies to mitigate lymphedema risk and enhance patient quality of life after surgery. Our study provides long-term insights into the potential benefits of ILR as a preventive measure against BCRL. The primary finding is that patients undergoing ILR had low rates of lymphedema, with only 7.0% of patients developing lymphedema during a mean follow-up of 23.1 months. Notably, Black or African American patients had significantly higher rates of lymphedema compared to White patients a finding that persisted in multivariable regression analysis. This is an alarming finding that warrants further investigation to better understand the underlying mechanisms and identify targeted interventions to mitigate these disparities.
    Since 2017, ILR has been widely adopted at our institution as a standard adjunct procedure for patients undergoing ALND, preventing the possibility of establishing a contemporary internal control group, as there are no comparable patients who underwent ALND without ILR during this period. Given the absence of a direct control group, we compared our findings to previously published pooled estimates of 10,774 patients undergoing ALND without ILR. Compared with pooled estimates from patients undergoing ALND without ILR, our ILR cohort had significantly lower lymphedema prevalence at all time points: 2.5% versus 16.5% at less than 12 months, 3.7% versus 24.6% between 12 and 24 months, and 7.0% versus 23.6% beyond 24 months postoperatively (p < 0.001).5 While these findings suggest an advantage in incorporating ILR into surgical planning for breast cancer patients requiring ALND, we are unable to adjust for differences in study design, patient populations, surgical techniques, and follow-up protocols as potential sources of bias. Therefore, our study does not provide a direct statistical comparison to historical data but rather an observational reference. Future prospective studies with standardized control groups are needed to better define the long-term impact of ILR on BCRL prevention.
    Our results align with previous studies that have highlighted the potential role of ILR in reducing BCRL incidence. A recent meta-analysis of 10 studies and 1,487 patients undergoing ALND found that only 7.9% of patients in the ILR group developed BCRL, compared with 20.8% in the control group, yielding a relative risk of 0.31.44 Another study evaluating 90 patients with a median follow-up of 17 months demonstrated a lymphedema rate of 9%, reinforcing the benefit of ILR in mitigating lymphedema risk.29 Additionally, our findings are consistent with a randomized controlled trial comparing outcomes of 72 patients who underwent ILR with 72 patients who had ALND alone, which reported a cumulative lymphedema incidence of 9.5% in the ILR group versus 32% in the non-ILR group.28 The relatively large sample size of 172 patients and the extended average follow-up period of 23 months in our study provide additional evidence supporting the long-term efficacy of ILR, contributing to the growing body of literature advocating for ILR as a preventive strategy for BCRL.
    Two studies have reported no benefit of ILR in preventing lymphedema.33,45 A possible reason for this discrepancy, and a general limitation to lymphedema literature, may be from inconsistency in defining the outcome “lymphedema.” Following axillary lymphadenectomy, lymphatic drainage from the upper extremity is altered, leading to postoperative swelling that may benefit from compression therapy. Patients may experience transient postoperative swelling that resolves over time.26,34 Defining lymphedema as any postoperative compression requirement or any swelling may overestimate clinically-relevant lymphedema rates. In our study, 78.2% of patients who used compression garments postoperatively did not meet the diagnostic criteria for lymphedema. Similarly, a randomized, prospective study by Coriddi et al. reported compression use in 30% of the ILR group at 24 months postoperativley.28 However, their primary outcome, lymphedema defined as 10% relative volume change of the extremity, showed a significantly lower incidence in the ILR group (9.5%) compared with the control group (32%).28 Lymphoscintigraphy is a valuable tool for assessing lymphatic function with a 96% sensitivity and 100% specificity.46 Obtaining lymphoscintigraphy at 3 years postoperatively may be ideal for comparing lymphatic function between groups.8 However, it is important to note that an abnormal lymphoscintigram in the presence of a normal physical examination may not be clinically significant.
    Despite the overall reduction in lymphedema rates, a subset of patients in our cohort still developed lymphedema following ILR. We analyzed various risk factors associated with lymphedema development—an area not extensively explored in previous studies. In our adjusted analyses, Black or African American race was associated with a sixfold higher likelihood of BCRL following ILR, indicating that certain patient characteristics may continue to influence lymphedema risk despite the protective benefit of ILR. This finding aligns with prior investigations demonstrating a racial disparity in lymphedema incidence.10,47–50 Montagna et al.47 reported that Black race was independently associated with nearly a fourfold higher risk of lymphedema following ALND, and Kwan et al.50 observed a twofold increased risk of BCRL among African American patients. Proposed mechanisms include dysregulated scarring and chronic inflammation in certain African-origin populations, which can lead to excessive connective tissue overgrowth and predispose patients to lymphatic compromise. Additionally, aberrant IL-4/IL-13 signalling pathways, commonly elevated in lymphedematous tissues and found at higher levels in Black individuals, may further exacerbate fibrosis and edema. While our findings suggest a significant disparity, it is important to acknowledge that our study may be underpowered to fully assess the true extent of this difference due to the limited number of Black or African American patients in our cohort. Despite this limitation, the observed racial disparity in lymphedema rates remains an alarming finding that warrants further investigation in larger, prospective studies. Collectively, these findings underscore the need to investigate the biologic underpinnings of lymphedema risk in Black or African American patients, particularly connective tissue reactivity and inflammatory regulation. Future efforts to elucidate these mechanisms and address potential disparities in access to care may guide targeted strategies to mitigate BCRL in this high-risk population.
    Our study has limitations. The nonrandomized, retrospective design may introduce selection bias and confounding factors, potentially limiting the generalizability of our results. However, conducting the study across six hospitals and involving multiple experienced microsurgeons and breast surgeons enhances the robustness of our findings. Additionally, the absence of a control group of patients who underwent ALND alone prevents direct comparative analyses. However, our institute has five microsurgeons who offer ILR. Immediate lymphatic reconstruction is essentially always offered during ALND. Therefore, a contemporary control group of patients undergoing ALND without ILR was not available. Alternatively, we compared lymphedema prevalence in our cohort with a large, pooled estimate of 10,774 patients. Additionally, whereas our study includes a relatively large cohort, the number of patients who developed lymphedema was limited. As a result, our regression models may be underpowered to detect significant differences in lymphedema rates across all subgroups. Furthermore, we did not utilize diagnostic measures, such as relative volume change, bioimpedance spectroscopy, or lymphography, to define lymphedema; instead, we employed two-point limb circumference measurements, previously reported as robust methods for identifying women at increased risk of developing permanent lymphedema.37

    Conclusions

    Immediate lymphatic reconstruction following ALND is associated with a significant reduction in the prevalence of BCRL compared with ALND alone, with notably lower rates observed across all time intervals. Despite this overall benefit, a subset of patients still developed lymphedema. Our data also indicate that Black or African American race remains a risk factor for BCRL even among patients undergoing ILR, highlighting persistent disparities in lymphedema outcomes. These findings underscore the multifactorial nature of BCRL and emphasize the need for further research to characterize its biologic and healthcare-related determinants, optimize patient selection for ILR, and implement strategies aimed at reducing treatment-related disparities.
    CFDr. Carla Fisher

    Dr. Carla Fisher

    Surgery · Breast Surgical Oncology
    Indiana University School of Medicine
    carla