Purpose: Breast radiotherapy (BrRT) often includes a "boost" to the lumpectomy bed. Selection of an appropriate electron energy and field size is based upon the dimensions and location of the surgical cavity. This study was performed to confirm our impression that ultrasound (US) inadequately defines the volume at risk compared to radiographic evaluation of surgical clips placed within the operative bed at the time of lumpectomy. METHODS AND MATERIALS Twenty-nine women treated with BrRT at our institution between 1996-1998 were prospectively identified as having surgical clips within the lumpectomy cavity. These patients underwent both US evaluation and radiographic evaluation of the clips to independently define two sets of transverse (T), longitudinal (L), and depth (D) measurements for each cavity. Volumes (V) were calculated for each method, assuming the cavity to be a box (V = T x L x D). Twenty-one women began BrRT following a median postoperative interval of 6 weeks (Group A), and 8 after 24 weeks (Group B) due to delivery of systemic therapy.
Results: Dimensions derived by US were smaller than the clip method in 22/29 of T, 25/29 of L, and 23/29 of D, or 80% of all linear measurements. A paired t-test demonstrated the difference between the methods to be statistically significant: T: p = 0.0004; L: p = 0.0001; D: p = 0.0004; and V: p = 0.0001. This underestimation by US did not fit any predictable pattern. Although the mean difference between methods (clips - US) was only 1.3, 1.9, and 1.1 cm for T, L, and D, respectively, differences ranged up to 5. 7, 9.2, and 5.5 cm for T, L, and D. The bias toward underestimation of V by US was significantly greater for Group B than Group A (p = 0. 03).
Conclusions: US significantly underestimates all three dimensions of the lumpectomy cavity and hence the volume at risk compared to radiographic evaluation of surgical clips. Breast-conserving surgery should include placement of clips at the margins of the lumpectomy cavity (superior, inferior, medial, lateral, and posterior) to aid in radiotherapy treatment planning. US should not be utilized to guide the design of the boost field as it will result in inappropriate selection of lower electron energies and smaller field sizes (geographical miss), particularly among patients who receive chemotherapy prior to breast radiotherapy.