Cervical re-injection of indocyanine green to improve sentinel lymph node detection in endometrial cancer

OBJECTIVES

To evaluate the role of cervical re-injection of indocyanine green (ICG) to increase the detection rate of sentinel lymph node (SLN) in patients with endometrial cancer (EC) who underwent robotic-assisted surgical staging.

METHODS

We retrospectively identified consecutive EC patients undergoing robotic-assisted staging with SLN biopsy at our Institution between June 2016 and April 2020. Patients were excluded if they had open abdominal surgical approach, neoadjuvant chemotherapy, and advanced stage [International Federation of Gynecology and Obstetrics (FIGO) stage III-IV] at diagnosis. According to our SLN protocol, in case of either unilateral or no SLN detection, we performed an ipsilateral or bilateral cervical re-injection of ICG.

RESULTS

In total, 251 patients meeting inclusion criteria were included in the analysis. At first injection, bilateral detection was achieved in 184 (73.3%), unilateral detection in 57 (22.7%), and no detection in 10 (4.0%) patients. Cervical re-injection was performed in 51 of 67 patients with failed bilateral mapping. After cervical re-injection, bilateral detection rate increased to 94.5% (222/235), while unilateral and no detection were 5.1% (12/235) and 0.4% (1/235), respectively.

CONCLUSIONS

Our results suggest that cervical re-injection of ICG, in case of failed bilateral mapping of SLN, brings about a significant improvement in SLN detection rates, therefore reducing the number of side-specific required lymphadenectomies.

Accuracy of transvaginal sonography and contrast-enhanced magnetic resonance-colonography for the presurgical staging of deep infiltrating endometriosis

OBJECTIVE

To investigate the accuracy of transvaginal sonography (TVS) and contrast-enhanced magnetic resonance-colonography (CE-MR-C) for the presurgical assessment of deep infiltrating endometriosis (DIE).

METHODS

Ninety women were enrolled prospectively for suspicion of DIE. All patients underwent TVS and CE-MR-C, with each operator blinded to the results of the other exam, before laparoscopy. The sites of DIE examined by both imaging techniques were: rectovaginal septum, pouch of Douglas, uterosacral ligaments, vesicouterine pouch, bowel, bladder and vagina. The presence of adhesions and the involvement of adnexa and of a previous abdominal scar, when there was clinical suspicion, were also evaluated. TVS and CE-MR-C findings were compared with laparoscopic and histological results.

RESULTS

Endometriosis was confirmed by laparoscopy in 95.6% (86/90) of cases. In 82.2% (74/90) of patients there was DIE. The global accuracy for TVS in the detection of DIE was 89.2%, sensitivity was 81.1%, specificity was 94.2%, positive predictive value was 89.6%, negative predictive value was 89.0%, the positive likelihood ratio was 13.9 and the negative likelihood ratio was 0.2. For CE-MR-C, these values were 87.2%, 71.1%, 97.1%, 93.7%, 84.6%, 24.4 and 0.3, respectively. CE-MR-C allowed diagnosis of all cases of bowel involvement; the accuracy for infiltration and stenosis was 100%. The accuracy of TVS for rectosigmoid nodules was 91.1% and that for infiltration was 88.9%.

CONCLUSIONS

Both TVS and CE-MR-C showed satisfactory results for the presurgical assessment of DIE. TVS appears to be a powerful, simple, feasible, cost-effective tool for preoperative staging of DIE. CE-MR-C is an 'X-ray free' technique, which could be reserved for cases with deep infiltrating rectosigmoid lesions and for the prediction of stenosis and involvement of the upper part of the colon and small intestine.