Sentinel lymph node identification by SPECT/CT in breast cancer

Abstract

Objective To evaluate the application of SPECT/CT with radioactive sulfur colloid (99Tcm-SC) in detecting sentinel lymph nodes (SLNs) in breast cancer patients. Methods According to the inclusion and exclusion criteria, a total of 118 breast cancer patients in the Affiliated Hospital of Qingdao University from April 2015 to August 2015 were enrolled in this retrospective study. All patients underwent preoperative 99Tcm-SC SPECT/CT combined with intraoperative blue dye tracing to detect SLNs. With the results of SPECT/CT plus blue dye tracing as the gold standard, the detection rate of SLNs by planar imaging, SPECT/CT or blue dye tracing alone was compared. All detected SLNs were sent to the pathologic examination. The detection rate of SLNs (based on cases) was compared using Fisher exact probability method, the detection rate of SLNs (based on node number) was compared using paired χ2 test, and the detection rate of pathologically positive SLNs by SPECT/CT and blue dye tracing was compared using paired χ2 test. Results With the results of SPECT/CT plus blue dye tracing as the gold standard, the detection rate of SLNs (based on cases) by planar imaging, SPECT/CT or blue dye tracing alone was 91.53%(108/118), 97.46%(115/118)and 97.46%(115/118) respectively, indicating a significant difference (P=0.020), and the detection rate by SPECT/CT or blue dye tracing was significantly higher than that by planar imaging(χ2=5.143, 5.143; P=0.023, 0.023). The detection rate of SLNs (based on node number) by planar imaging, SPECT/CT and blue dye tracing was 61.72% (158/256), 96.88% (248/256) and 95.70% (245/256) respectively, indicating a significant difference (χ2=158.072, P<0.001). The detection rates by SPECT/CT and blue dye tracing was significantly higher than that by plane imaging (χ2=90.000, 87.000; both P<0.001). The detection rate of SLNs in axillary region by planar imaging, SPECT/CT and blue dye tracing was 64.44% (154/239), 96.65% (231/239) and 100.00% (239/239) respectively, indicating a significant difference (χ2=163.312, P<0.001). The detection rate of SLNs in axillary region by SPECT/CT and blue dye tracing was significantly higher than that by planar imaging(χ2=77.000, 85.000; both P<0.001), while the detection rate of SLNs in axillary region by blue dye tracing was significantly higher than that by by SPECT/CT (χ2=6.125, P=0.013). The detection rate of SLNs in non-axillary region by planar imaging, SPECT/CT and blue dye tracing was 4/17, 17/17 and 6/17 respectively, indicating a significant difference(χ2=23.139, P<0.001). The detection rate of SLNs in non-axillary region by SPECT/CT was significantly higher than that by planar imaging and blue dye tracing (χ2=11.077, 9.091; P=0.001, 0.003). The pathologically positive rate of SLNs detected by SPECT/CT and blue dye tracing was 24.22% (62/256) and 21.09% (54/256) respectively, indicating no significant difference (χ2= 3.500, P = 0.061). For axillary lymph nodes, eight SLNs which were not detected by SPECT/CT were found by blue dye tracing, including 3 with macrometastasis and 5 without metastasis in the pathological examination. For non-axillary lymph nodes, 11 SLNs (3 internal mammary nodes, 5 interpectoral lymph nodes, 3 intramammary nodes) which were not detected by blue dye tracing were found by SPECT/CT, including 9 with micrometastasis and 2 with macrometastasis in pathological examination. Conclusions SPECT/CT can provide guidance for accurate detection of SLNs and precise positioning of SLN biopsy. Furthermore, it can improve the detection rate of pathologically positive SLNs, especially for non-axillary SLNs, superior to blue dye tracing. Key words: Breast neoplasms; Sentinel lymph node; Radionuclide imaging