Cervical versus Utero-Ovarian Ligament Injection of the Tracer for the Pelvic Sentinel Lymph Node Mapping in Gynecologic Oncology: A Prospective Observational Study

Objective: In gynecologic oncology, the consolidated injection site for the pelvic sentinel lymph node mapping is the cervix. However, in apparent early-stage ovarian cancer, current trials map the pelvic area injecting the tracers in the utero-ovarian ligament. A different injection site is proposed based on a possible different lymphatic ovarian drainage through the utero-ovarian ligament and uterus and consequently a different pelvic sentinel lymph node mapping than the cervix; however, this was never proven. On that basis, this study aimed to investigate whether injecting the tracer in the utero-ovarian ligament map the same or a different pelvic sentinel lymph node than the cervix. Design: A prospective observational study was conducted. Methods: All consecutive women undergoing primary surgery for gynecologic malignancy with planned pelvic sentinel lymph node mapping were enrolled. The cervical injection was performed at 3 and 9 o'clock injecting indocyanine green. Bilateral utero-ovarian ligament injection was performed by injecting methylene blue. The probability of detecting the same sentinel lymph node (concordance rate) and the probability of detecting a different sentinel lymph node (discordance rate) in each hemipelvis were compared using McNemar's exact test. Results: Out of 36 hemipelvis (18 patients), the overall detection rate with cervical indocyanine green injection was 86.1% (31/36) versus 52.8% (19/36) with utero-ovarian ligament methylene blue injection (p = 0.0004). Indocyanine green and methylene blue identified the same sentinel lymph node in all hemipelvis when sentinel lymph node mapping was obtained by both dyes (19/19; concordance rate 100%). No different or additional sentinel lymph nodes were identified by one of the two dyes (0/19; discordance rate 0%). The probability of detecting the same pelvic sentinel lymph node by the two injection sites was significantly higher than the probability of detecting a different pelvic sentinel lymph node (p < 0.0001). Limitations: The use of two different dyes with known different sentinel lymph node mapping performance impedes to evaluate differences in detection rate attributable to the two injection sites, limiting the comparison only in hemipelvis in which both dyes identified a sentinel lymph node. Conclusions: The cervix and utero-ovarian ligament are two sites for the tracer injection that have a higher probability of detecting the same than a different pelvic sentinel lymph node. Given the higher technical challenge in performing the utero-ovarian ligament injection, these results suggest that the injection of the tracer into the cervix could be a more reliable alternative for the pelvic sentinel lymph node mapping in ovarian cancer.