Data collection and population.
This retrospective study was compliant with the “CMN 20 de noviembre ISSSTE” institutional review board.
Informed consent was not required.
All consecutive mammography and breast US reports obtained between January 1,
2015,
and December 31,
2017,
were identified in the institutional database.
All breast examinations with short-term follow-up (BI-RADS category 3) were included in the study.
Biopsy for both palpable and nonpalpable masses (BI-RADS categories 4 and 5) as well as masses known to be malignant (BI-RADS category 6) were not included the study.
Medical records were reviewed to determine imaging finding outcomes,
including pathologic examination findings and outcome of these masses.
Breast images were not re-reviewed.
During the study,
591 patients’ examinations revealed lessions assessed as BI-RADS 3.
Re-classification was assessed at the short and mid-term follow-up.
Patients lost-of follow-up were excluded from final analysis.
Imaging and interpretation.
Breast US examinations were performed by staff radiologists (20 years of experience using Aloka Alpha 6 equipment with a linear 5–12- MHz transducer.
Mammography examinations were performed using Hologic Selenia digital series equipment.
Imaging interpretation was based on BIRADS Atlas 5th ed.
The decision to recommend follow-up imaging included mammographic and US findings,
and the decision to perform biopsy in the following examinations was based on the finding most suggestive of possible malignancy.
BI-RADS assessment categories were assigned to each case by using ACR definitions.
Follow-up recommendations and outcomes.
When short-interval imaging was recommended (BI-RADS category 3),
the patient was asked to return to a 6-month follow-up.
If the lesion was stable,
a second 6-month follow-up (12 months after the initial examination) was typically recommended.
At 12-month evaluation,
if the mass was stable,
the examiner categorized the result as BI-RADS category 3 or 2,
with recommendation for follow-up in 12 months.
Despite a BI-RADS category 3 recommendations,
the patient or referring physician might have decided to perform biopsy.
For the purposes of this study,
the latter scenario is defined as physician and/or patient preference.
Biopsies were performed either under US or stereotactic guidance depending on main imaging features.
After pathology results were available,
concordance with imaging was confirmed,
and the referring physician and patient were contacted with results and recommendations.
Discordant results were discussed at routine weekly radiology-pathology conferences.
Statistical analysis:
Descriptive statistics were used to report frequencies of BIRADS 4 and 2 lesions among patients initially classified as BIRADS 3,
and expressed as percentages.
Univariate analysis was performed using Chi-square/ Fisher tests to assess the impact risk factors on malignancy outcomes.
SAS,
version 9.4; SAS,
Cary,
NC) was for analysis.
A P value of less than 0.05 was considered significant difference.