Sentinel lymph node mapping and biopsy for breast cancer at a rural-based university medical center: initial experience with intraparenchymal and intradermal injection routes.

Background: Recent data suggests that intradermal (ID) injection for sentinel lymph node (SLN) mapping and biopsy in breast cancer is as effective and reproducible as intraparenchymal (IP) injection. The aim of this study was to review our initial experience with IP and ID injection for SLN mapping and biopsy at our rural-based university medical center.

Methods: From January 4, 1999 to January 5, 2001, 113 of 165 patients with breast cancer underwent attempted SLN mapping and biopsy by either IP (n=63) or ID (n=50) injection. Selection of the IP versus ID injection route was non-randomized and based on surgeon preference. Success of SLN localization was examined.

Results: SLN localization was successful in 82% of IP and 100% of ID for radioisotope (p=0.001), 69% of IP and 92% of ID for blue dye (p=0.002), and 90% of IP and 100% of ID (p=0.024) for radioisotope and blue dye. Identical comparisons made after excluding the first 10 cases, 20 cases, and 30 cases from each injection group showed that the percentage of cases in each group in which the SLN localized changed minimally; however, some of the resultant p values eventually lost statistical significance.

Conclusions: SLN localization was more successful by ID injection than by IP injection, thus favoring utilization of the ID injection route. The eventual loss of statistical power in some of the comparisons with increasing numbers of initial cases excluded may reflect differential learning curves of the two injection techniques; however, this may simply be a reflection of decreasing sample size used in each subsequent analysis. A prospective randomized trial comparing the IP and ID injection route may be warranted.