Abstract

Aims—In recent years there has been increased use of core biopsy for the preoperative diagnosis of screen detected and symptomatic breast lesions. The aim of this study was to compare the quality assessment parameters for preoperative diagnosis by fine needle aspiration cytology (FNAC) before conversion to core biopsy with those for core biopsy after conversion in screening and symptomatic practice. Accuracy of typing and grading of tumours on core biopsy was assessed.

Methods—Correlation of FNAC (C1–5) and core biopsy (B1–5) results (total of 1768 cases) with subsequent available resection histology was performed for 473 FNAC samples in 1997/98, 349 core biopsies in 1998/99 performed in symptomatic practice, for 561 FNAC samples in 1997/98, and 385 core biopsies in 1998/99 performed in screening. Quality assessment parameters were calculated using the methodology detailed in the National Health Service Breast Screening Programme guidelines for cytology practice.

Results—Increased absolute and complete specificity, lower inadequate rates, and lower suspicious rates were found for core biopsy compared with FNAC in both symptomatic and screening practice. Typing of tumours was attempted in 86.7% of core biopsies in symptomatic practice and was accurate in 93.6% (132 of 141 where type was stated). Grading of tumours was attempted in 63.5% of invasive carcinomas, with the provisional grade on core biopsy being confirmed on later histology in 75% of grade 1 cases, in 70% of grade 2 cases, and in 86% of grade 3 cases. No case provisionally graded as 1 was subsequently found to be grade 3 and no provisionally grade 3 case was found to be grade 1.

Conclusion—Conversion to core biopsy for the preoperative diagnosis of breast lesions increases specificity and reduces inadequate and suspicious rates. Grading and typing of tumours and assessment of oestrogen receptor status by immunocytochemistry is also possible in core biopsy, thereby increasing diagnostic information available when considering treatment options.