Objective: The purpose of this study was to evaluate intraoperative imprint cytology (IC) for the detection of sentinel node (SN) involvement in patients with cervical cancer.
Methods: Thirty-six consecutive patients with cervical cancer underwent a laparoscopic SN procedure with intraoperative IC, followed by complete laparoscopic pelvic lymphadenectomy, with or without laparoscopic para-aortic lymphadenectomy. The SN was bisected and both cut surfaces were applied to the surface of glass slides. Permanent sections were stained with H&E and immunohistochemical methods. The IC results were compared with the final histological results.
Results: At least one SN (mean: 2 SN per patient, range: 1-5) was detected in 34 of the 36 patients. Eight patients (22.2%) had a total of 12 metastatic SNs (four macrometastatic, five micrometastatic, and three with isolated tumor cells). No false-negative results of SN biopsy were obtained. Only one metastasis was identified by IC. No false-positive findings were obtained with IC. The overall sensitivity, specificity, accuracy, and positive and negative predictive values of IC were 8.3%, 100%, 85.7%, 100%, and 85.5%, respectively.
Conclusion: These results suggest that intraoperative imprint cytology of sentinel nodes is unreliable in patients with cervical cancer.