Jacky Luiten

Recall and outcome of screen‐detected microcalcifications of mammography screening | 77 5 Trends in diagnostic work‐up of microcalcifications Of the 3556 women recalled for suspicious microcalcifications, 617 (17.4%) underwent additional imaging (additional mammographic views, tomosynthesis, breast US, breast MRI, or any combination of these modalities) as the only diagnostic procedure (Table 5.2). Stereotactic core needle biopsy (SCNB) was the most frequently used type of biopsy, and a total of 2484 SCNB procedures were performed (Table 5.2). The SCNB rate per 1000 screening examinations increased from 0.4 (150/351,009) during the SFM period to 1.2 (109/90,128) during the transition period ( p <0.001), remaining stable in the FFDM period (1.2; 451/176,519). The proportion of women who underwent SCNB for microcalcifications increased from 0% (0/51) in 1997–1998 to 77.6% (528/680) in 2015‐2016 ( p <0.001). A total of 1111 suspicious microcalcifications proved to be (pre)malignant (31.2%; 1111/3556) and consisted of 747 DCIS (67.2%) and 364 invasive cancers (32.8%) (Table 5.3). The preoperative confirmation of DCIS with percutaneous biopsy increased over the years, from 5.9% (1/17) in 1997‐1998 to 91.8% (135/147) in 2015‐2016 ( p <0.001) (Table 5.3). Trends in surgical treatment of DCIS Of all 747 women with DCIS lesions that presented with suspicious calcifications, 561 (75.1%) were treated with breast conserving surgery and 173 (23.2%) with mastectomy. The remaining thirteen women (1.7%) did not undergo surgical treatment (Table 5.3). The lowest proportion of breast conserving surgery for screen‐detected DCIS was observed in women in the first cohort of 1997‐1998 (59%; 10/17) and subsequently varied between 70% (28/40) and 81% (38/47) afterward. DCIS characteristics The median DCIS size was 19 mm (range, 1‐130 mm). Of the 747 DCIS lesions, 393 (52.6%) were high grade, 233 (31.2%) were intermediate grade, and 116 (15.5%) were low grade (Table 5.3). This distribution of low, intermediate, and high grade DCIS did not change during the whole study period (SFM vs. FFDM period, p =0.36). The distributions of DCIS lesions according to histologic grade per 1000 screening examinations in the SFM and FFDM periods were 0.1 (41/351,009) and 0.3 (102/376,519), respectively, for low grade DCIS ( p <0.001); 0.2 (58/351,009) and 0.5 (192/376,519) for intermediate grade DCIS ( p <0.001); and 0.5 (175/351,009) and 0.6 (241/376,519; p =0.01) for high grade DCIS (Figure 5.4). Following SCNB,