Clinical Utility of MicroPure US Imaging for Breast Microcalcifications.

Purpose: To evaluate the performance of MicroPure US imaging to detect and characterize microcalcifications. Materials and
Methods: A total of 171 lesions with suspicious microcalcifications seen on mammography and B-mode US were included and simultaneously evaluated using MicroPure US imaging. The size of microcalcifications was divided into small (punctate, amorphous, fine pleomorphic, and fine linear) and large (coarse heterogeneous), and the extent was divided into narrow (grouped) and wide (others). MicroPure US imaging visibility was divided into four types based on the number of microcalcifications on the two images: B > M (more on B-mode), B = M (similar), B < M (more on MicroPure), and negative. Triple pairwise comparison was used to evaluate the imaging features according to the MicroPure US imaging visibility.
Results: Among the 171 lesions examined, 157 lesions (91.8%) were detected by MicroPure US imaging. The proportion of Breast Imaging Reporting and Data System (BI-RADS) category 4A was significantly higher in the MicroPure positive group, and that of category 4B was significantly higher in the MicroPure negative group (p = 0.035). The other imaging features did not differ. Among the positive MicroPure subgroups, all features showed no significant difference.
Conclusion: MicroPure US imaging demonstrated 91.8% positivity in detecting microcalcifications on B-mode US. MicroPure US imaging visibility correlated with the BI-RADS category of microcalcifications. Copyrights

INTRODUCTION

Microcalcifications in both screening and diagnostic mammography are important diagnostic clues for breast cancer. Microcalcification is highly associated with ductal carcinoma in situ, and approximately 20% of invasive cancer is associated with microcalcifications (12). Stereotactic biopsy is an accurate biopsy method for suspicious microcalcifications detected on mammography (3). Advances in US technology have enabled radiologists to identify and localize microcalcifications (4). US guidance is preferred in lesions that can be biopsied with both stereotactic or US guidance and is currently considered an effective and reliable alternative to stereotactic biopsy in such cases (45). This is because US guidance has the advantages of simplicity and cost-effectiveness, radiation-proof procedure, and flexible needle insertion with real-time observation (6). However, microcalcifications without mass are difficult to detect on US because they are small and buried in the echogenic, fibroglandular breast tissue (78). In addition, not all “hyperechoic foci” on US are microcalcifications, and collagen fibers and other changes might simulate microcalcifications (9).

MicroPure™ (Canon Medical Systems, Tustin, CA, USA) is a new US imaging processing technique to highlight microcalcifications by combining non-linear imaging and speckle suppression with a constant false alarm rate filter. This filter extracts only isolated high-brightness echoes against a heterogeneous background (1011). MicroPure US imaging marks suspected calcifications as white spots in a blue overlay image (12). A few studies have demonstrated superior detection of larger numbers of microcalcifications with MicroPure US imaging than with B-mode US (10111314).

To the best of our knowledge, no studies have analyzed detection and visibility capabilities of MicroPure US imaging based on imaging features. The purpose of this study was to evaluate the performance of MicroPure US imaging for detecting microcalcifications seen on mammography and B-mode US and to identify characteristics of microcalcifications detected with MicroPure US imaging.

MATERIALS AND METHODS

SUBJECTS

The Institutional Review Board of our hospital approved this retrospective study, and the requirement for informed consent was waived (IRB No. KC19RESI0686).

This retrospective study performed from September 2017 to March 2019 included 171 cases. Mammographic reporting and US examination were performed by one of three radiologists with a breast subspeciality (5–19 years of experience). Breast Imaging Reporting and Data System (BI-RADS) category 4A or higher microcalcifications on mammography and that were also shown on B-mode US were included. Using mammography as a reference, the region of interest was scanned with B-mode US (4–18 MHz probe, Aplio i700, Canon Medical Systems). When echogenic foci on B mode US were matched with mammographic calcifications, a simultaneous MicroPure US image was obtained as a still image (MicroPure, Aplio i700, Canon Medical Systems). Pathologic diagnosis was performed with US-guided core-needle biopsy (n = 161), US-guided vacuum-assisted biopsy (n = 5), and stereotactic vacuum-assisted biopsy (n = 5). Specimen mammography after biopsy (n = 87) or pathologic report confirmed the presence of microcalcifications. With biopsy or surgery, final pathology was benign (n = 29) or malignant (n = 142) (Table 1).

Table 1

Pathologic Results (n = 171)

	Number of Lesions
Invasive breast cancer	107
Ductal carcinoma in situ	35
Atypical ductal hyperplasia	6
Lobular carcinoma in situ	1
Mucocele-like lesion	1
Fibrocystic change	14
Fibroadenoma	4
Sclerosing adenosis	3

IMAGE ANALYSIS

Medical records were reviewed, and images of mammography, B-mode US, and MicroPure US were reviewed with consensus by two radiologists (one with 14 years of experience in breast imaging and the other, a senior radiology resident) who were blinded to the pathologic results.

Morphology and distribution of the microcalcifications on mammography were evaluated using BI-RADS. Size and extent of microcalcifications were divided into two groups. Regarding size, the ‘small’ group was composed of punctate, amorphous, fine pleomorphic, and fine linear/linear branches smaller than 0.5 mm in size, and the ‘large’ group was composed of coarse heterogeneous calcifications. Regarding extent, the ‘narrow’ group was composed of grouped calcifications less than 2 cm in spread, and ‘the wide’ group was composed of regional, segmental, and linear distributions.

US images were evaluated whether the microcalcifications were associated with mass- or non-mass-like lesions and whether background echotexture was homogeneous or heterogeneous.

MicroPure US imaging was considered positive for microcalcification when a bright white spot was visible in an area of echogenic foci on B-mode. MicroPure US imaging visibility was divided into four types based on number of calcifications on two images: B > M (more calcifications on B-mode than on MicroPure), B = M (similar numbers), B < M (more on MicroPure than on B-mode), and negative (no microcalcifications on MicroPure) (Fig. 1).

Fig. 1

Representative type of MicroPure visibility.

B-mode US (left) shows echogenic foci matched with microcalcifications detected on mammography (right). MicroPure US imaging (middle) shows bright dots.

A. In B > M type, MicroPure US imaging shows fewer bright dots than echogenic foci on B-mode.

B. In B = M type, MicroPure US imaging and B-mode US show similar patterns.

C. In B < M type, MicroPure US imaging shows more bright dots than echogenic foci on B-mode US.

D. B-mode US (left) shows echogenic foci matched with microcalcifications detected on mammography (right). MicroPure US imaging (middle) shows no bright dots.

STATISTICAL ANALYSIS

To assess pairwise differences between the two groups, Fisher’s exact test was employed, because the negative MicroPure group comprised a small number of cells. The ANOVA test was used for triple pairwise comparison among MicroPure-positive groups. A p value < 0.05 was considered significant. Statistical analysis was performed using SAS Version 9.4 (SAS Institute, Cary, NC, USA).

RESULTS

Mammographic and US findings of 171 lesions with microcalcifications are summarized in Table 2. There was 7.6% of large calcifications (coarse heterogeneous calcifications [n = 13, 7.6%]) and 92.4% of small calcifications (punctate [n = 6, 3.5%], amorphous [n = 59, 34.5%], fine pleomorphic [n = 67, 39.2%] and fine linear [n = 26, 15.2%]). Extent according to distribution was wide in 59% of the calcifications (regional, segmental, linear distribution) and narrow in 41% (grouped distribution).

Table 2

Mammographic and Ultrasonographic Findings of Microcalcifications (n = 171)

		Total
Mammographic abnormalities
	Calcification	98 (57.3)
	Calcification with mass	73 (42.7)
Morphology of calcifications
	Punctate	6 (3.5)
	Amorphous	59 (34.5)
	Coarse heterogeneous	13 (7.6)
	Fine pleomorphic	67 (39.2)
	Fine linear/linear branching	26 (15.2)
Size of calcifications
	Large	13 (7.6)
	Small	158 (92.4)
Distribution of calcifications
	Grouped	70 (41.0)
	Regional	45 (26.3)
	Segmental	53 (31.0)
	Linear	3 (1.7)
Extent of calcifications
	Wide	101 (59.0)
	Narrow	70 (41.0)
Mammographic category (BI-RADS)
	C4A	12 (7.0)
	C4B	66 (38.9)
	C4C	48 (28.0)
	C5	45 (26.3)
Ultrasonographic abnormalities
	Echogenic foci	43 (25.1)
	Echogenic foci with mass	102 (59.7)
	Echogenic foci with nonmass	26 (15.2)
Background echotexture
	Heterogeneous	95 (55.6)
	Homogeneous	76 (44.4)
Pathology
	Benign	29 (17.0)
	Malignant	142 (83.0)

Numbers in parentheses are percentage.

BI-RADS = Breast Imaging Reporting and Data System

Among 171 lesions, 91.8% (157/171) were detected on MicroPure US imaging (Table 3). MicroPure US imaging did not detect calcifications in 14 cases (8.1%). Most of the negative group were calcifications without mass on mammography (9/14, 64.3%), grouped distribution (10/14, 71.4%), small size (13/14, 92.9%), narrow extent (10/14, 71.4%), and malignancy (13/14, 92.9%). Grouped distribution was more frequently shown in MicroPure negative group than in MicroPure positive group (71.4% vs. 38.2%, respectively), but there was no significant difference (p = 0.062). The proportion of C4A was significantly higher in the MicroPure positive group and that of C4B was significantly higher in the MicroPure negative group (p = 0.035) (Figs. 2, 3). Mammographic abnormalities, morphology, size, distribution and extent of calcifications, US abnormalities, background echotexture and pathology were not different between MicroPure positive and negative group.

Table 3

Mammographic and Ultrasonographic Findings according to MicroPure US Imaging Visibility

		Positive (n = 157)	Negative (n = 14)	p-Value
Mammographic abnormalities				0.084
	Calcification	89 (56.7)	9 (64.3)
	Calcification with mass	68 (43.3)	5 (35.7)
Morphology of calcifications				0.064
	Punctate	6 (3.8)	0 (0.0)
	Amorphous	54 (34.4)	5 (35.7)
	Coarse heterogeneous	12 (7.6)	1 (7.1)
	Fine pleomorphic	62 (39.5)	5 (35.7)
	Fine linear/linear Branching	23 (14.6)	3 (21.4)
Size of calcifications				0.108
	Large	66 (42)	6 (42.9)
	Small	91 (58)	8 (57.1)
Distribution of calcifications				0.062
	Grouped	60 (38.2)	10 (71.4)
	Regional	44 (28)	1 (7.1)
	Segmental	50 (31.8)	3 (21.4)
	Linear	3 (1.9)	0 (0.0)
Extent of calcifications				0.153
	Wide	97 (61.8)	4 (28.6)
	Narrow	60 (38.2)	10 (71.4)
Mammographic category (BI-RADS)				0.035
	C4A	12 (7.6)	0 (0.0)
	C4B	59 (37.6)	7 (50.0)
	C4C	45 (28.7)	3 (21.4)
	C5	41 (26.1)	4 (28.6)
Ultrasonographic abnormalities				0.345
	Echogenic foci	38 (24.2)	5 (35.7)
	Echogenic foci with mass or nonmass	119 (75.8)	9 (64.3)
Background echotexture				0.084
	Heterogeneous	88 (56.1)	7 (50.0)
	Homogeneous	69 (43.9)	7 (50.0)
Pathology				0.389
	Benign	28 (17.8)	1 (7.1)
	Malignant	129 (82.2)	13 (92.9)

Numbers in parentheses are percentage.

BI-RADS = Breast Imaging Reporting and Data System

Fig. 2

A 41-year-old female with suspicious microcalcifications and positive MicroPure findings.

A. Magnification view shows regional, amorphous microcalcifications, with Breast Imaging Reporting and Data System category 4A.

B. B-mode US (left) shows a few echogenic foci of microcalcifications (arrowheads) and MicroPure US imaging (right) shows one bright dot of microcalcification (arrowhead).

C. Specimen mammography after US-guided core needle biopsy confirms the presence of microcalcifications and fibrocystic change.

Fig. 3

A 56-year-old female with suspicious microcalcifications and negative MicroPure findings.

A. Magnification view shows grouped, fine pleomorphic microcalcifications, with Breast Imaging Reporting and Data System category 4B.

B. B-mode US (left) shows few echogenic foci of microcalcifications (arrowheads). MicroPure US imaging shows no bright dots.

C. Specimen mammography after US-guided core needle biopsy confirms the presence of microcalcifications and invasive breast carcinoma.

All lesions according to MicroPure US imaging visibility were divided into B > M (n = 92, 53.8%), B = M (n = 51, 29.8%), B < M (n = 14, 8.1%), and negative (n = 14, 8.1%) (Tables 3, 4). All imaging features including distribution of calcifications and pathology showed no difference among the three MicroPure US imaging visible groups.

Table 4

Comparison of Imaging Features and Pathology among the Groups Stratified by MicroPure US Imaging Visibility

		B > M (n = 92)	B = M (n = 51)	B < M (n = 14)	p-Value
Mammographic abnormalities					0.205
	Calcification	49 (53.3)	29 (56.9)	11 (78.6)
	Calcification with mass	43 (46.7)	22 (43.1)	3 (21.4)
Size of calcifications					0.972
	Large	8 (8.6)	3 (5.8)	1 (7.1)
	Small	84 (91.4)	47 (94.2)	13 (92.9)
Extent of calcifications					0.087
	Wide	63 (68.5)	28 (54.9)	6 (42.9)
	Narrow	29 (31.5)	23 (45.1)	8 (57.1)
Mammographic category (BI-RADS)					0.172
	C4A	6 (6.5)	4 (7.8)	2 (14.3)
	C4B	28 (30.4)	23 (45.1)	8 (57.1)
	C4C	28 (30.4)	14 (27.5)	3 (21.4)
	C5	30 (32.6)	10 (19.6)	1 (7.1)
Ultrasonographic abnormalities					0.376
	Echogenic foci	22 (23.9)	13 (25.5)	3 (21.4)
	Echogenic foci with mass	52 (56.5)	31 (60.8)	11 (78.6)
	Echogenic foci with nonmass	18 (19.6)	7 (13.7)	0 (0.0)
Background echotexture					0.461
	Heterogeneous	55 (59.8)	25 (49.0)	8 (57.1)
	Homogeneous	37 (40.2)	26 (51.0)	6 (42.9)
Pathology					0.083
	Benign	12 (13.0)	11 (21.6)	5 (35.7)
	Malignant	80 (87.0)	40 (78.4)	9 (64.3)

Numbers in parentheses are percentage. B > M: MicroPure shows fewer microcalcifications as bright dots than does B-mode. B = M: MicroPure and B-mode show microcalcifications in similar numbers. B < M: MicroPure shows more microcalcifications as bright spots than does B-mode.

BI-RADS = Breast Imaging Reporting and Data System

DISCUSSION

We evaluated the effectiveness of a new ultrasound image processing technique, MicroPure US imaging, in detecting microcalcifications in female with suspicious microcalcifications on mammography and B-mode US.

MicroPure US imaging demonstrated positivity in 91.8% of the microcalcifications detected on both mammography and B-mode US. In previous studies, both B-mode and MicroPure US imaging showed 100% detection positivity in diffuse microcalcifications (n = 20 patients) (13), grouped microcalcifications (n = 10 patients) (10), and microcalcifications (n = 20 surgical specimens) (11). Another study with 70 cases of pathologically proven microcalcifications revealed a higher detection rate for MicroPure US imaging than for B-mode (100% vs. 71.4%). In the present study, MicroPure US imaging did not detect microcalcifications in 8.2% of cases (14/171). We analyzed the characteristics of imaging features in MicroPure negative case, and only the BI-RADS category was meaningful. The proportion of of category 4B was significantly higher in the MicroPure negative group and that of category 4A was significantly higher in the MicroPure positive group in this study. On B-mode US, it was more difficult to identify isolated microcalcification than to find microcalcifications within a hypoechoic mass because of the lack of contrast between the parenchyma and the microcalcifications (13). However, associated US abnormalities did not differ between the MicroPure US imaging visible group and the negative group in this study.

Study results showed that 91.8% of microcalcifications were visible with MicroPure US imaging and B > M type was the most frequent (53.8%), followed by B = M (29.8%) and B < M types (8.1%). In previous studies, clinicians saw more calcifications with MicroPure US imaging than with B-mode in all cases (1113). One study showed B < M (60%) and B = M (40%) in 10 cases (10). The present study showed the B < M group to be smallest in proportion, which was inconsistent with previous studies (101113). The B < M group was possibly related to MicroPure US imaging artifacts due to Cooper’s ligaments (13) and fibrosis and hyaline change (14). The presence of artifacts had no effect on the accuracy of identification of microcalcifications (15).

Biopsy with US guidance is preferred to stereotactic biopsy for suspicious microcalcifications on mammography and US due to several advantages (456). Accurate US-guided biopsy for suspicious microcalcifications needs two steps confirming precise correlation of mammographic and US findings. First, additional mammography is obtained after placing the adhesive marker on the skin at the area of suspicious microcalcifications on B-mode. Second, specimen mammography is performed after biopsy. MicroPure US imaging improved visualization of microcalcifications and allowed physicians to perform biopsy of microcalcifications (1016). When microcalcifications are found on B-mode and are also shown as bright dots on MicroPure US imaging, the confidence of detection on US is increased. Considering the high detection positivity of MicroPure, it can replace additional mammography after skin marking in US-guided biopsy for microcalcifications detected both on B-mode and MicroPure US imaging (Fig. 4).

Fig. 4

Schematic diagram of conventional US-guided biopsy and the newly proposed method using MicroPure imaging.

Conventional US-guided biopsy for suspicious microcalcifications consists of four processes to confirm the precise correlation between mammography and US and the presence of microcalcifications: detection with B mode, additional mammography after skin marking, biopsy, and specimen mammography. Considering its high detection positivity, MicroPure US imaging might become an alternative to additional mammography after skin marking.

We think that MicroPure US imaging will be useful for radiologists who do not have much experience in breast US for the detection of microcalcifications. Also MicroPure US imaging will help the guidance of needle and be useful in reducing the biopsy procedure time through improved visualization of microcalcifications.

This study had some limitations in that it was a retrospective study from a single institution with a limited number of cases. There might have been selection bias, and the results might not be applicable to the general population. Including only cases where mammographic calcifications were visible on B-mode US, this study did not evaluate and compare the detection rates of B-mode and MicroPure US imaging for mammographic suspicious microcalcifications.

Finally, still images of B-mode and MicroPure US were reviewed instead of video images. Therefore, limited area was included for evaluation and artifacts and microcalcifictions were not properly distinguished.

In conclusion, MicroPure US imaging showed comparable positivity to B-mode US in detecting microcalcifications. BI-RADS category of microcalcifications significantly differed according MicroPure US imaging visibility.