[Critical study of our initial experience of 993 sentinel node biopsies for breast surgery].

Objective: Identification of sentinel node (SN) involvement predictive factors, non-sentinel node involvement predictive factors, selective prognosis of each group of patients by study of breast surgery cases with sentinel node sampling.

Methods: Prospective monocentric registering of 993 sentinel node samples routinely taken between January 2001 and October 2005, covering technical aspects of detection (colorimetric and radio-isotope), pathological results (serial sections 5 Mmicro thick prior to staining hematoxylin-eosine-saffron and if necessary, by immune histochemistry cytokeratine high molecular weight), therapeutics and follow-up (average period: 32 months (3-69).

Results: Seven hundred and sixteen patients (72.1%) were free of sentinel node involvement. Among positive sentinel node patients (27.9%), 14.5% presented macrometastasis, 11% micrometastasis and 2.4% isolated tumor cells (CTI). Sentinel node involvement risk factors included: related to clinical features, age (2 years younger in the micrometastatic group compared to the macrometastatic group); related to tumor caracteristics, size (12.15 mm for the negative SN group, 15.4 mm for the micrometastatic group and 16.25 mm for the macrometastatic group), grading (a majority of grade I encountered with micrometastasis versus macrometastasis) and multifocality (macrometastasis SN associated with multilocular tumor in 77.8% cases, micro metastasis SN in 22.2% cases and negative SN in 6.7% cases). Predictive factors do not differ for micro- or macrometastasic involvement. Among features concerning secondary axillary dissection, 47.1% (66/140) were positive with a macrometastatic SN, 12.1% (13/107) with micrometastic SN. Predictive factors of positive secondary axillary dissection were tumor size, grading, micrometastasis size and micrometastasis multifocality. With a 32 months mean follow-up, the positive micrometastasis sub-group (with or without positive secondary axillary dissection) expressed one only metastatic recurrence (0.9%); on the contrary, three patients (2.1%) issued from the macrometastatic SN group, expressed metastatic recurrence. One only local axillary recurrence (0.14%) occurred among negative SN (717 cases); no axillary recurrence occurred among the 30 patients without secondary axillary dissection (CTI [22 cases], micrometastatic SN group [5 cases] and macrometastatc group [3 cases]).

Conclusion: First, 72.1% of T0 or T1 tumors, avoid adverse axillary dissection effects. Second, micrometastatic involvement predictive factors do not differ from macrometastatic ones and those of positive secondary axillary dissection among micrometastatic SN do not appear clearly : the risk of axillary recurrence is low: at the very most, it seems possible to propose a safe guideline, avoiding secondary axillary dissection only for selected group of lower risk patients: tumoral size < 10 mm, grade I, monocentric SN involvement. Third, it is not possible to differentiate a selective prognosis between negative, CTI, micrometastatic and macrometastatic SN subgroups probably because of a short follow-up. Fourth, teaching through companionship is fully valided by the secondary minimal rate of axillary recurrence.