Breast cancer-related lymphedema after axillary lymph node dissection: does early postoperative prediction model work?

PURPOSE: Early detection and timely intervention demonstrate the greatest promise of reducing the incidence of late-stage lymphedema in breast cancer patients undergoing axillary lymph node dissection (ALND). A nomogram was developed for predicting the risk of lymphedema (LE) in patients with ALND. This study's aim was to test the early postoperative prediction model for the diagnosis of clinical and subclinical LE after ALND. METHODS: Patients requiring ALND were identified preoperatively through our LE program database. Measurements using metered tape with bioimpedance spectroscopy (L-Dex U400) were obtained preoperatively (n = 180) and at 3-6-month intervals postoperatively. The 5-year probability of LE after ALND was calculated using the Cleveland Clinic Risk Calculator. The discrimination of the nomogram was assessed by calculating the area under (AUC) the receiver operating characteristic curve. RESULTS: LE was present in 36.1% (n = 65) of 180 patients with ALND. Of these 65 patients, 22 (12.2%) had clinical LE and 43 (23.9%) had subclinical LE. Statistical analyses showed significant differences in BMI and receipt of radiotherapy between patients with and without LE (p = 0.03 and p = 0.01, respectively). AUC was 0.601, 0.614, and 0.600 for the nomogram using any LE, clinical LE, and subclinical LE patients, respectively. CONCLUSIONS: The recently created prediction model for the diagnosis of LE in ALND is not accurate in predicting who will develop clinical or subclinical LE. Periodic monitoring of women with ALND is the most effective method to aid in reducing clinical LE incidence through early detection and timely intervention of LE.