The usefulness of a preoperative compact imager, a hand-held gamma-camera for breast cancer sentinel node biopsy: final results of a prospective double-blind, clinical study
Khaldoun Kerrou, Stéphanie Pitre, Charles Coutant, Roman Rouzier, Pierre-Yves Ancel, Cécile Lebeaux, Virginie Huchet, Françoise Montravers, Odile Pascal, Marie-Alix Duval, Françoise Lefebvre, Laurent Menard, Serge Uzan, Yves Charon, Emmanuel Barranger, Khaldoun Kerrou, Stéphanie Pitre, Charles Coutant, Roman Rouzier, Pierre-Yves Ancel, Cécile Lebeaux, Virginie Huchet, Françoise Montravers, Odile Pascal, Marie-Alix Duval, Françoise Lefebvre, Laurent Menard, Serge Uzan, Yves Charon, Emmanuel Barranger
Abstract
The aim of this study was to compare the effectiveness of a hand-held preoperative compact imager (POCI) camera with conventional lymphoscintigraphy using a γ-camera for sentinel lymph node (SLN) detection in breast cancer.
Methods: The main objective was to demonstrate the noninferiority of the POCI relative to conventional lymphoscintigraphy and to compare the number of SLNs detected by the 2 imaging devices. Our study, a clinical prospective, double-blind, noninferiority study, planned to include 200 patients with early breast cancer and started in January 2006. A standard SLN protocol (4 periareolar injections of 37 MBq of (99m)Tc-nanocolloids, 2 h before lymphoscintigraphy) was performed preoperatively using a conventional γ-camera and then the POCI camera. Scans were obtained by 2 different nuclear medicine physicians unaware of each other's results. The day after, in the operating room, the surgeon, after receiving the previous results, used the counting probe for surgical SLN biopsy. The number and localization of axillary SLNs obtained by lymphoscintigraphy and the POCI and the duration of the whole procedure were determined.
Results: Among the 162 patients included, 138 were evaluable. The POCI detected more SLNs than did lymphoscintigraphy in 50 patients (36%), the same number of in 54 patients (39%), and fewer SLNs in 34 patients (25%), representing 84 (61%) discordant pairs. The noninferiority of preoperative compact imaging of axillary SLNs numbers was found to be statistically significant (95% confidence interval, 30%-52%, P = 0.025) using the McNemar test. The duration of acquisition was shorter using the POCI (<10 min in 84% [n = 117] of patients; mean, 7.5 ± 3.3 min) than lymphoscintigraphy (13% [n = 18] of patients; mean, 15.7 ± 3.4 min), with P < 0.001 using the McNemar test for paired proportions.
Conclusion: Preoperative compact imaging using a hand-held camera was able to predict the number and localization of breast cancer SLNs and was not inferior to conventional lymphoscintigraphy in this study. Further studies will determine whether preoperative compact imaging could replace lymphoscintigraphy, especially in surgical centers without an on-site nuclear medicine department.
Source: PubMed