Subpleural injection of tracer improves detection of mediastinal sentinel lymph nodes in non-small cell lung cancer

Abstract

The advantages and disadvantages of various tracer injection protocols for sentinel lymph node (SLN) mapping have been extensively discussed in relation to breast and gastric cancer. But no such discussion has taken place in relation to SLN mapping in non-small cell lung cancer. We therefore studied the effect of two tracer injection protocols on SLN mapping in patients with non-small cell lung cancer; of particular interest was the relationship between subpleural tracer injection and identification of mediastinal SLNs. A quadrant injection group (n=49) received 1.6 ml of ferucarbotran by peritumoral quadrant injection after thoracotomy. In the subpleural injection group, the same amount of ferucarbotran was injected into the peritumoral quadrants plus the subpleural region (n=27). SLNs were then detected intraoperatively by measuring the magnetic force within lymph nodes using a hand-held magnetometer. After completing the SLN mapping, lobectomy and hilar and mediastinal lymph node dissection was performed. The incidence of mediastinal SLNs was significantly higher in the subpleural injection group (45.4%) than in the quadrant injection group (14.6%) (p=0.007). Moreover, nominal logistic regression analysis revealed subpleural injection to be a significant independent factor contributing to detection of mediastinal SLNs (p=0.024, odds ratio 5.26). In the quadrant injection group, mediastinal lymph node metastasis was detected in two patients thought to have nonmetastatic parenchymal SLNs. By contrast, there were no false-negative cases in the subpleural injection group. Subpleural tracer injection significantly improves detection of mediastinal SLNs in non-small cell lung cancer.