Double reading mammograms in the symptomatic setting

Symptomatic breast clinics in the UK provide specialised assessment and treatment for patients with breast-related symptoms, including one-stop and follow-up clinics for patients under surveillance. The NHS Breast Screening Programme (NHSBSP) invites asymptomatic women between the ages of 50 and 70 years for breast cancer screening every three years. The NHSBSP employs double reading; however, its application in the symptomatic setting is debated. This study examines the impact of single versus double reader mammograms on cancer detection rates and arbitration time delays.

A retrospective analysis was performed between 2021 and 2023 of one-stop and surveillance clinic cases. Only cases in which one reader recommended recall (BI-RADS M3-M5) versus no recall (BI-RADS M1-M2) were considered discordant and required arbitration. Out of 9405 patients, 164 met the inclusion criteria (1.74% arbitrated cases). Information on reader decisions and patient outcomes were used to explore the impact of arbitration in each cohort.

The average delay from the first read to arbitration was 12.7 days. The greatest impact was observed within the surveillance cohort. Of the 44 surveillance cases recalled by the first reader, 38 were downgraded to no recall following arbitration. Conversely, following arbitration, the second reader recalled 13 cases, 8 of which proceeded to biopsy: 4 were malignant and 4 were benign. In the symptomatic cohort, the second reader recalled 37 of 75 cases following arbitration, 18 of which underwent biopsy. This resulted in the detection of 4 additional malignancies, 5 cases of unknown malignant potential and 9 benign results.

Double reading with arbitration improves cancer detection rates, particularly in surveillance cases. Although arbitration introduced a 12.7-day delay, the second reader identified eight malignancies and five B3 lesions, underscoring the value of this approach.