Full-field digital mammography compared to screen film mammography in the prevalent round of a population-based screening programme: the Vestfold County Study

Abstract

The purpose of the study was to compare the performance of full-field digital mammography (FFDM) with soft-copy reading to screen film mammography (SFM) used during the first prevalent 2-year round of population-based screening. A total of 18,239 women aged 50-69 years were screened with FFDM as part of the Norwegian Breast Cancer Screening Programme (NBCSP). Process indicators were compared to data from 324,763 women screened with SFM using the common national database of the NBCSP. The cancer detection rates were 0.77% (140/18,239) for FFDM and 0.65% (2,105/324,763) for SFM (p = 0.058). For ductal carcinoma in situ (DCIS) alone, the results were: FFDM 0.21% (38/18,239) compared to SFM 0.11% (343/324,763) (p < 0.001). Recall rates due to positive mammography were for FFDM 4.09% (746/18,239), while for SFM 4.16% (13,520/324,764) (p = 0.645), due to technically insufficient imaging: FFDM 0.22% (40/18,239) versus SFM 0.61% (1,993/324,763) (p < 0.001). The positive predictive value (PPV) in the FFDM group was 16.6% (140/843), while 13.5% (2,105/15,537) for SFM (p = 0.014). No statistically significant differences were recorded concerning histological morphology, tumour size, or lymph node involvement. In conclusion FFDM had a significantly higher detection rate for DCIS than SFM. For invasive cancers no difference was seen. FFDM also had a significantly higher PPV and a significantly lower technical recall rate.

Is full field digital mammography equivalent in performance to screen film mammography? UK perspective

“…Use of CAD could lead to more sensitive detection of micro-calcifications by Karssemeijer et al [18] . The fact that previous studies [5,6] without CAD also reported increased detection of micro-calcifications, suggest that this result should not be attributed to CAD alone. Another factor may be the distribution of patients between the first screening round and subsequent rounds.…”

“…Oslo 2 study [4] , showed significantly higher cancer detection rate with FFDM in women between 45 and 69 years of age (5.9 versus 3.8 per 1000 (p = 0.002). Studies [5,6,8,17] done in the context of routine screening mammograms showed similar results. Del Turco et al [5] and Vigeland et al [6] compared SFM and FFDM in a local population based programme and found statistically higher detection rate for FFDM, which was marked for cancers presenting with microcalcification.…”

“…Studies [5,6,8,17] done in the context of routine screening mammograms showed similar results. Del Turco et al [5] and Vigeland et al [6] compared SFM and FFDM in a local population based programme and found statistically higher detection rate for FFDM, which was marked for cancers presenting with microcalcification. The Norwegian study by Vigeland et al [6] was done over a nine year period and read by different radiologists in 18 different counties.…”

“…Del Turco et al [5] and Vigeland et al [6] compared SFM and FFDM in a local population based programme and found statistically higher detection rate for FFDM, which was marked for cancers presenting with microcalcification. The Norwegian study by Vigeland et al [6] was done over a nine year period and read by different radiologists in 18 different counties. By contrast, in our study, experienced readers remained constant over the period of the study and although more microcalcification was detected in FFDM group in our study, this was not found to be statistically significant.…”

“…There has been replacement of screen-film mammography (SFM) with full field digital mammography (FFDM) in many centres because of consistent image quality of digital images, ability to post-process digital images and improved storage and transmission capabilities [3] . A number of studies have been conducted to compare screen-film with digital mammography and results have shown that digital imaging is as good as SFM in population based screening practices [4][5][6][7][8] , although most of these studies rely on surrogate end points such as cancer detection rate, recall rate and tumour characteristics at diagnosis. Few studies have been done within an operational national screening programme [9][10][11][12] .…”