Patients and lesions
Our prospective study included a series of 151 patients who consulted at Pierre and Marie Curie Anti Oncology Center,
between January 2016 and September 2017.
Among these patients,
187 breast lesions detected by B-mode US were analyzed with strain elastography and color Doppler US.
The exclusion criteria were patients with a personal history of breast cancer and patients under radiotherapy or neoadjuvant chemotherapy for neighboring lesions.
B-mode Ultrasound
All sonographic examinations were performed by using ultrasound machine systems equipped with a 13-6 MHz linear transducer (Hi-Vision system,
Hitachi Medical).The mass classified as BI-RADS category 3 (probably benign),
4a (low suspicion for malignancy),
4b (moderate suspicion for malignancy),
4c (high suspicion for malignancy),
or 5 (highly suggestive of malignancy) [11]. BI-RADS category 3 was assigned to oval,
circumscribed,
isoechoic masses with parallel orientation.
BI-RADS category 4a,
4b,
4c or 5 was assigned according to the number of suspicious outcomes (Fig.
1) [10].
Its largest measurement in millimeters (mm) was retained after measuring the mass in three axes.
Color Doppler ultrasound
Color Doppler US was performed by using standardized parameter settings (pulse repetition frequency between 700–1000 Hz,
wall filter as low as possible [50–100 Hz],
appropriate algorithm to remove motion artifacts,
maximum gain [85%–90%],
medium persistence,
and box without angulation).Vascularity for a lesion was interpreted according to the BI-RADS atlas [11],
classified as no vascularity ,
vessels in rim and/or internal vascularity (Fig.
2).
Strain Elastography
Elasticity Scores
Elastographic images of all lesions were evaluated according to the 5-point elasticity scoring system suggested by Itoh and al [12].
A score of 1 corresponded to green in the entire hypoechoic tumor on elastography; 2 corresponded to a mosaic pattern of green, blue and deformity of most of the tumor; 3 corresponded to blue in the central part and green in the peripheral part of the tumor; 4 corresponded to blue in the entire hypoechoic tumor; and 5 corresponded to blue in the entire hypoechoic tumor and its surrounding area.
According to the criteria of Itoh and al [12],
tumors of scores 1,
2,
and 3 were categorized as benign,
and those of scores 4 and 5 were categorized as malignant (Fig.
3).
Strain Ratio
Calculation of the strain ratio was based on the comparison of the average strain measured in the lesion with the fat tissue.
Using proprietary software on the ultrasound machine,
the average strain of the lesion was determined by selecting a representative region of interest from lesion (L) and was expressed as ''L''.
A corresponding region of interest into the fat tissue (F) was then selected,
and the average strain was expressed as ''F''.
The resultant strain ratio was calculated according to the equation strain ratio = F/L,
which reflected the property of stiffness of the lesion (Fig.
4).
Histology
The histology of all the masses was established with US-guided 14-gauge automated gun core biopsy.
US-CNB results were classified into FA or PT.
Pathologists with specialized breast training divided into three categories according to stomal cellularity as defined by Jacobs TW and al [1] : ''Mildly and moderately cellular stroma" regarding FA and ''markedly cellular stroma'' for PT (Fig.
5).
Statistical analysis
Age,
lesion size on B-mode US,
Strain Elastography features and vascularity on the color Doppler US of FA and PT were compared between FA and PT groups.
All numeric data were represented the mean ± standard deviation (SD).
Non-numeric data were represented by the number of lesions,
the percentage and were analyzed using the chi-square test or Fisher's exact test.
Spearman’s correlation coefficient was calculated between strain ratio and pathological groups.
In the analysis of diagnostic performance for differentiating FA from PT,
positive test results for PT were defined as BI-RADS category 4a or higher for B-mode US and high vascularity (equal to vessels in rim and internal vascularity) for color Doppler US,
respectively.
In terms of SE features,
positive results of quantitative elasticity values (strain ratio ) were determined using the cut-off values obtained according to the Youden index [18].
The sensitivities,
specificities,
accuracies,
positive predictive values (PPV),
and negative predictive values (NPV) of B-mode US,
SE features (strain ratio) and color Doppler US were calculated to differentiate FA from PT.
The area under the curve (AUC) was also calculated after construction of the receiver operating characteristic (ROC) curve.
The AUC,
sensitivity,
specificity,
accuracy,
PPV,
and NPV of B-mode US alone were compared with three combined tests; B-mode US with SE,
B-mode US with color Doppler US,
B-mode US with SE and color Doppler US respectively with McNemar's test and the chi-square test. Statistical significance was accepted when P values < 0.05.
Statistical analysis were performed by using the Statistical Package for the Social Sciences (SPSS version 23 for Mac,
Inc.,
Chicago,
IL,
USA).