Morbidity of sentinel node biopsy: relationship between number of excised lymph nodes and patient perceptions of lymphedema.

Sentinel lymph node biopsy (SLNB) is associated with reduced morbidity, although lymphedema remains a significant complication. Previously, we found no association between number of excised lymph nodes (LNs) and measured lymphedema in SLNB patients. In this analysis, we examined the relationship between number of LNs excised during SLNB and patient-perceived lymphedema. A total of 600 women who underwent SLNB for breast cancer were prospectively studied. Measured lymphedema was evaluated by circumferential bilateral upper-extremity measurements taken preoperatively and 3-8 years postoperatively. Patient-perceived lymphedema was evaluated by interview at follow-up. The relationship between lymphedema, total LNs excised, and clinicopathologic variables was assessed with Fisher's exact test, Wilcoxon rank-sum test, kappa statistic, and McNemar's test. At a median of 5 years, 18 (3%) patients reported perceived lymphedema. More LNs were excised in patients with perceived lymphedema compared with those without (median, 5.5 vs. 3; p = 0.01). Only 6 of 18 women with perceived lymphedema had objectively measured lymphedema (kappa = 0.22). Patients with numbness more likely reported perceived lymphedema (p = 0.03) and had more LNs excised (p = 0.02). Women with surgery on the nondominant axilla were less likely to perceive arm swelling, regardless of the presence of measured lymphedema. After SLNB alone, patient-perceived lymphedema is uncommon, but its prevalence increases with more LNs excised. There is poor agreement between patient perceptions and objective measures. Our data suggest that factors other than limb enlargement, such as sensory nerve injury resulting from retrieval of more LNs and laterality of surgery, may play a significant role in patient perception of lymphedema after SLNB.