The Oslo II Study randomly assigned women to screening by digital mammography (n = 6,944) versus SFM (n = 16,985) with soft-copy double reading by experienced radiologists. Recall and cancer detection rates were higher for digital mammography, but there was no difference in PPV or incidence of interval cancers.
A study in a single screening center in the Netherlands compared women (aged 50-75 years) attending a population-based screening program who were screened on a new full-field digital mammography (FFDM) unit (that included computer-aided detection [CAD]) with women being screened by SFM. For a period of 5 years, a total of 311,082 screening examinations were done by SFM and 56,518 by FFDM. The groups were assembled without obvious bias but without randomization. The recall rate was higher in the FFDM group (4.41% vs. 2.32% at first screen and 1.70% vs. 1.17% at subsequent screens, both P < .001). There was no statistically significant difference in the detection of invasive breast cancer (4.9 per 1,000 SFM vs. 5.4 per 1,000 FFDM at first screen [P = .46] and 4 per 1,000 SFM vs. 4 per 1,000 FFDM [P = .96] at subsequent screens between the groups). There was higher detection of DCIS in the FFDM group (2.2 per 1,000 FFDM vs. 1.2 per 1,000 SFM [P = .015] at first screen and 1.2 per 1,000 FFDM vs. 0.8 per 1,000 SFM [P = .007] at subsequent screens). Most of this increased detection of DCIS appears to be caused by increased detection of clustered microcalcifications by FFDM compared with SFM.
A review of ten controlled studies of various designs found that, overall, the literature supports an increase in breast cancer detection (combining invasive cancer and DCIS), and that the evidence is mixed concerning which modality is associated with higher recall rates.
The performance of mammography is very different in the United States as compared with the Netherlands. Specifically, the recall rates are much higher, and with similar cancer detection rates, the PPVs are much lower. Thus, the impact of digital mammography with CAD versus SFM in the United States may be different.
CAD systems are designed to assist radiologists in reading mammograms. The goal is to help identify suspicious regions such as clustered microcalcifications and masses. The use of CAD systems increases sensitivity but decreases specificity. Several CAD systems are in use. One large population-based study comparing recall rates and breast cancer detection rates before and after the introduction of CAD systems questions their utility; there was no change in either rate.[52,54] Another large study noted an increase in recall rate, no improvement in cancer detection rate, and an increased detection of DCIS compared with invasive cancers. Because no mortality studies have been conducted, the impact of CAD on breast cancer mortality is uncertain. CAD systems seem to increase detection of DCIS more than invasive breast cancers.