National Outcomes of Prophylactic Lymphovenous Bypass during Axillary Lymph Node Dissection

Background Breast cancer treatment, including axillary lymph node excision, radiation, and chemotherapy, can cause upper extremity lymphedema, increasing morbidity and health care costs. Institutions increasingly perform prophylactic lymphovenous bypass (LVB) at the time of axillary lymph node dissection (ALND) to reduce the risk of lymphedema but reports of complications are lacking. We examine records from the American College of Surgeons (ACS) National Surgery Quality Improvement Program (NSQIP) database to examine the safety of these procedures. Methods Procedures involving ALND from 2013 to 2019 were extracted from the NSQIP database. Patients who simultaneously underwent procedures with the Current Procedural Terminology (CPT) codes 38999 (other procedures of the lymphatic system), 35201 (repair of blood vessel), or 38308 (lymphangiotomy) formed the prophylactic LVB group. Patients in the LVB and non-LVB groups were compared for differences in demographics and 30-day postoperative complications including unplanned reoperation, deep vein thrombosis (DVT), wound dehiscence, and surgical site infection. Subgroup analysis was performed, controlling for extent of breast surgery and reconstruction. Multivariate logistic regression was performed to identify predictors of reoperation. Results The ALND without LVB group contained 45,057 patients, and the ALND with LVB group contained 255 (0.6%). Overall, the LVB group was associated with increased operative time (288 vs. 147 minutes, p < 0.001) and length of stay (1.7 vs. 1.3 days, p < 0.001). In patients with concurrent mastectomy without immediate reconstruction, the LVB group had a higher rate of DVTs (3.0 vs. 0.2%, p = 0.009). Reoperation, wound infection, and dehiscence rates did not differ across subgroups. Multivariate logistic regression showed that LVB was not a predictor of reoperations. Conclusion Prophylactic LVB at time of ALND is a generally safe and well-tolerated procedure and is not associated with increased reoperations or wound complications. Although only four patients in the LVB group had DVTs, this was a significantly higher rate than in the non-LVB group and warrants further investigation.