Audit of performance of needle core biopsy diagnoses of screen detected breast lesions

Maysa E. El-Sayed, Emad A. Rakha, Jacquie Reed, Andrew H. S. Lee, Andrew J. Evans, Ian O. Ellis

    Research output: Contribution to journalArticlepeer-review

    35 Citations (Scopus)


    Breast needle core biopsy (NCB) is now a standard diagnostic procedure in the triple assessment of screen detected breast lesions. Therefore, it is important to provide robust and up-to-date data on the performance of NCB in the screening setting. However, previous studies of NCB have suffered from either limitation in the number of assessed cases or included a mix of symptomatic and screen detected breast lesions. in this study, we have evaluated the performance of a large series of uniformly assessed NCBs of screen detected lesions (20001 cases) over a period of 10 years (1997-2007). our results showed a gradual increase in the number of NCBs and an improvement of their performance over the period of the study; absolute sensitivity increased from 84.9% to 96,4% and complete sensitivity increased from 90.9% to 99.7%. There was also a gradual reduction in the number of surgical interventions after benign (B2) and negative (B1) NCB diagnoses. Our study provides data showing variance from the suggested thresholds for the measures of performance of NCB in the United Kingdom which could be used to provide updated evidence-based thresholds for assessment of performance of NCB diagnosis use in the assessment of breast cancer screen detected lesions in the UK and elsewhere. (C) 2008 Elsevier Ltd. All rights reserved.

    Original languageEnglish
    Pages (from-to)2580-2586
    Number of pages7
    JournalEuropean Journal of Cancer
    Issue number17
    Publication statusPublished - Nov 2008


    • Screen detected breast lesions
    • Needle core biopsy
    • Quality assurance
    • CANCER


    Dive into the research topics of 'Audit of performance of needle core biopsy diagnoses of screen detected breast lesions'. Together they form a unique fingerprint.

    Cite this