Patient population:
This single-center prospective study was approved by the institutional review board (IRB) and patients’ consents were obtained for all patients.
Between May 2011 and October 2014,
all the consecutive patients admitted in our center for multifocal or larger than 3 cm DCIS,
and therefore candidates for a mastectomy,
were enrolled in the Carcinome In-Situ et Plaque Aréolo-Mamelonnaire (CISPAM) protocol (age 33-70,
mean 50.1).
The mean follow up after surgery was 40.3 months (quartiles 25-75%: 31- 47 months).
CISPAM protocol:
The inclusion criteria in the CISPAM protocol were multifocal or larger than 3 cm DCIS,
with a normal aspect of the NAC.
The included patients had to undergo both a DM,
a tomosynthesis and a breast MRI.
The TND was measured on all available procedures and reported on Table 1.
Patients with a minimal TND ≥ 10 mm (on either one of these images) underwent a NSM (using previously described technique (8)),
or a SSM otherwise,
both with immediate breast reconstruction (either with prosthetic implant,
latissimus dorsal or Deep Inferior Epigastric Perforator flaps).
No intraoperative frozen sections were taken.
The top and bottom of the resected retro-areolar tissue were marked with suture threads,
so that the pathologist could measure the TND (1), and report all tumor extensions to the NAC.
It is noteworthy that the pathologist was blind to the imaging results.
If the final histology demonstrated NAC involvement or tumor-NAC distance of less than 10 mm,
a secondary surgical procedure with resection of the NAC would be performed.
Imaging protocols and data:
For all patients,
missing images were due to a special recruitment of the patients: some patients were sent from other hospitals,
often with at least a DM (not always found in the local patients files) and it was decided not to repeat these procedures.
Digital mammographies and Tomosyntheses were acquired in our center in 17/17,
and 13/17 (76.5%) patients,
respectively (Inspiration PRIME®,
Medical Solutions,
Erlangen,
Germany),
with either/or cranial-caudal (CC),
medial-lateral (ML) or medial-lateral oblique (MLO) views (for both techniques,
in the exception of MLO views,
only for DM).
The examiner measured on all available views the minimum distance between suspect microcalcifications or masses and the NAC,
and retained the smallest for each procedure and each patient.
3T breasts MRI were acquired in our center in 17/17 patients (MAGNETOM Verio®,
Siemens Medical Solutions,
Erlangen,
Germany) with a 16-channel breast coil.
For each patient,
we analyzed the subtracted imaged from the dynamic contrast-enhancement (CE) (with 0.5mmol/ml single dose of gadoteric acid) on 3D T1-weighted gradient echo images with 0.8 mm isotropic voxels.
TND were measured for each suspect mass or non-mass CE.
Long-term Follow-up
All patients were followed by the surgical oncology team,
including outpatient consultations with radiographic imaging (DM,
TS and MRI) every 6 months for the first two years and annually to look for evidence of recurrence of the disease.
Assessment also included secondary procedures,
complications,
and other post-operative surgical or medical managements.
Statistical analysis
For each patient; the pathological report was the gold standard test.
Statistical analysis was performed using Microsoft Office Excel version 2010 (Microsoft,
Redmond,
WA,
USA).