Keywords
breast - intervention - biopsy - vacuum-assisted biopsy
Introduction
Breast cancer incidence has progressively increased over the years, becoming the leading
cause of cancer in females worldwide.[1] Breast abnormalities are currently being assessed using the “triple test,” a comprehensive
approach encompassing clinical breast examination, imaging, and histopathological
correlation. Despite a multipronged approach of various breast imaging modalities,
lesions deemed as indeterminate or suspicious will still require a histological correlation,
thereby rendering surgical excision as the gold standard for breast abnormalities.
However, the associated cost, psychological burden, and duration of hospital stay
for open surgical biopsies are high. As a result, since decades, less invasive alternative
techniques such as “core needle biopsy” (CNB) whether image guided or nonguided, has
been incorporated into standard evaluation of breast lesions requiring histological
evaluation.[2]
[3]
[4]
[5]
[6] CNB may be plagued by histologic underestimation and false-negative diagnoses, especially
in smaller and complex lesions.[7]
[8]
[9] These caveats are further reinforced by the need for larger tissue volume imperative
for complete histopathological assessment, including analysis of molecular subtype,
tumor grade, receptor status, and genetic profile to guide further management. As
redressal, vacuum-assisted breast biopsy (VABB), a form of CNB powered by vacuum suction,
allowing contiguous retrieval of larger core samples, without the need for needle
reinsertion with a collateral benefit of vacuum evacuation of postbiopsy blood products
was developed.[10]
[11]
[12] First introduced in 1995, VABB has been accepted as an alternative method to CNB
to diagnose breast lesions with high sensitivity and specificity.[13]
[14]
[15]
[16] Hematoma formation is the most commonly associated complication with this sampling
technique.[17]
This study is a retrospective analysis of all consecutive, VABBs from March 2021 to
February 2022 at a tertiary cancer care center to evaluate safety and efficacy of
this sampling technique with respect to needle gauge and guiding modality.
Materials and Methods
Ethics Committee approval of a retrospective study with waiver of consent was granted
by the Institutional Review Board. Patient demographics and lesion data are shown
in [Table 1].
Table 1
Demographic and lesion data
|
Patients
|
60
|
|
Age (y), mean (range)
|
47 (22–70)
|
|
Gender
|
Female (n = 60)
|
|
Location in breast
|
|
Upper outer
|
24
|
|
Upper inner
|
11
|
|
Central, diffuse
|
13, 3
|
|
Lower outer
|
7
|
|
Lower inner
|
2
|
Patient Selection
The morphology of the findings during mammography (MG) and/or ultrasonography (US)
were interpreted and categorized as per the American College of Radiology Breast Imaging
Reporting and Data System (ACR BI-RADS) version 5.0.[18] According to standard lexicon recommendations, biopsy was considered for all categories
4 and 5 lesions, while for category 3 lesions, biopsy was performed at the discretion
of the referring physician. Subcentimeter-sized lesions and those with a complex morphology
(aggregation of ducts; initial nonrepresentative or discordant previous biopsy) were
planned for VABB. The decision of US versus stereotactic (MG) guidance was case specific,
considering the lesion's size, microcalcifications, and location in the breast. If
a lesion was detected by both MG and US, then, for noncalcific lesions, ultrasound
VABB (U-VABB) was preferred due to greater flexibility in needle placement, maneuvering,
and visibility of the procedure performed. For lesions containing microcalcifications,
and visible on US, the latter was still preferred for sampling a sonographically visible
(possibly invasive) component followed by specimen mammogram ([Fig. 1]). MG-only-detected suspicious microcalcifications, asymmetry, and architectural
distortion were targeted by stereotactic VABB (S-VABB). None of the patients were
on anticoagulants/antiplatelets and none of the patients were sedated. Patient demographics
and lesion distribution are summarized in [Table 1].
Fig. 1 A 46-year-old woman with a history of nipple discharge from right breast underwent
mammography. (A) Craniocaudal (CC) view showed regional distribution of amorphous calcification with
associated density. Targeted ultrasound revealed an ill-defined hypoechoic lesion
with multiple echogenic foci representative of microcalcifications (not shown). (B) Ultrasound image showing microcalcification (arrow) within the needle trough. (C) CC view showing post-VABB changes and clip marker in situ. Histopathology revealed
fibrocystic change, apocrine metaplasia, duct hyperplasia without atypia, and microcalcifications.
Patient is on routine surveillance.
Equipment
US was performed using a linear transducer (LA3–16A) (EVO RS80; Samsung Healthcare,
Seoul, South Korea), and MG was performed on Senographe Pristina (GE Healthcare, Milwaukee,
Wisconsin, United States) with an integrated stereotaxy facility. 10G and 7G vacuum-assisted
needle probes compatible with its dedicated vacuum device were used (EnCor Aspire;
BARD, Murray Hill, New Jersey, United States) for performing the biopsies. Written
informed consent was obtained from all patients.
Procedure
Patients were placed in the supine position for U-VABB, while for S-VABB, patients
were comfortably seated upright. Biopsy track and skin was infiltrated with 2% lidocaine
creating a wheal, followed by instillation of lidocaine–adrenaline (1:200,000) along
the track and surrounding the lesion to minimize bleeding.
In U-VABB, needle probe was positioned either juxta-superior/inferior or within the
lesion depending on its proximity to chest wall or skin, and its size, such that the
needle trough is epicentered along the lesion ([Fig. 2]).
Fig. 2 A 72-year-old woman, treated case of triple negative breast cancer 20 years ago,
on routine follow-up. Mammography (MG) showed a faint new density in the inner aspect.
A small 5-mm irregular hypoechoic lesion was seen on ultrasonography (US), and confirmed
to represent the corresponding MG-detected lesion with a mammogram after placement
of a (A) skin marker (triangle) and intralesional contrast instillation (B). US images showing (C) hypoechoic lesion (arrow) and the (D) needle along its inferior aspect (arrow head). Histopathology revealed invasive
ductal carcinoma.
In S-VABB, needle position was confirmed before sample retrieval by acquiring an additional
set of paired images after needle insertion, followed by sample acquisition. For lesions
with microcalcifications, specimen radiographs were acquired before concluding the
procedure to confirm their presence ([Fig. 3]).
Fig. 3 A 45-year-old woman with a family history of breast cancer underwent screening mammography.
(A) Biopsy planning craniocaudal view of left breast showed pleomorphic calcification
in segmental distribution. (B) Needle at target site. (C) Specimen radiograph confirming microcalcification in biopsy cores. Histopathology
revealed ductal carcinoma in situ, which was further confirmed on subsequent surgery.
Localizing marker (clip) was placed at the biopsy site in cases of near-complete excision
or in cases where the target was a single group of microcalcifications.
Post-VABB, cold compression was applied longitudinally encompassing the site of incision,
trajectory, and the lesion for ∼10 to 15 minutes until no oozing was observed, followed
by antiseptic dressing, hematoma volume assessment, and application of an elastic
compression bandage around the chest. Postprocedure hematoma volume was measured on
US immediately postbiopsy (day 0), after 24 hours (day 1), and on the 7th to 10th
day. Complications were categorized according to the “Society of Interventional Radiology”
(SIR) adverse event classification system.[19] No literature could be found on volume of hematoma considered as significant in
breast, and so for the purpose of this study, postprocedure hematoma volume of more
than 20 mL was defined as significant. All patients were contacted via telephone on
day 3 to inquire about local site discomfort, ooze, purulent discharge, and fever.
Analysis
Diagnostic yield (DY) was analyzed as per the SIR standards[20] where DY is defined as percentage of biopsies that result in a diagnosis. Histopathology
of the final surgical specimen or biopsy sample in that order of preference was considered
as the gold standard. Safety was evaluated by recording the “SIR-classified” complications
as minor (A, B) or major (C, D, E, F).[19] Pain level was recorded using the visual analog scale (VAS) ranging from 0 (no pain)
to 10 (worst pain experienced). Histologic underestimation was considered when “ductal
carcinoma in situ” (DCIS) on VABB was upgraded to invasive carcinoma on surgical excision.
Statistical analysis for postprocedure hematoma volume and its association with needle
gauge (10G vs. 7G) or guiding modality (US vs. MG) was performed using the Mann–Whitney's
U test and the results were considered statistically significant for p-value < 0.05. Statistical analysis was performed using SPSS (the Statistical Package
for Social Sciences), IBM Corp, released 2012, IBM SPSS Statistics for Windows, Version
21.0. Armonk, New York, United States: IBM Corp, and RStudio, version 1.1463, RStudio
Inc.
Results
Results are summarized in [Table 2]. Sixty patients underwent VABB between March 2021 and February 2022, of which one
was a therapeutic excision (n = 1) on the patient's request ([Fig. 4]) and the rest (n = 59) were diagnostic. [Table 3] gives an overview of lesions categorized as per ACR BI-RADS and their percentage
of malignancy on histopathology, with most lesions classified as ACR BI-RADS category
4. Histopathology reports of all patients confirmed 100% DY with one case of underestimation
(4%) where U-VABB yielded DCIS; however, invasive ductal carcinoma was found on surgery.
In terms of safety, no major complications, requiring hospital stay or intervention
were encountered. Minor complication of postprocedural hematoma not requiring nominal
therapy or intervention was observed in 42 cases, while 18 patients showed no measurable
hematoma immediately after the procedure or on subsequent follow-up imaging. Average
postprocedure hematoma volume on day 0 was ∼1.8 mL (0.6–2.5) in S-VABB, 2 mL (1–3)
in U-VABB, 1.6 mL (0.9–3) with 10G needle, and 2 mL (1.3–3) with 7G needle ([Figs. 5] and [6]). Of the 42 cases, hematoma volumes were nonmeasurable by day 7 in 31 cases, and
there was no significant correlation between postprocedure hematoma volume and needle
gauge (p = 0.2) or imaging guiding modality (p = 0.4). Procedure was well tolerated with most patients (62%) experiencing only mild
pain ([Table 4]).
Fig. 4 A 42-year-old woman with clinically palpable right breast lump. Ultrasonography revealed
a well-defined oval hypoechoic mass (solid arrow). Vacuum-assisted breast biopsy (VABB)
was performed with intent of excision. (A) VABB needle probe is seen along the inferior aspect of the mass (arrowheads). (B) Gradual decrease in lesion size (split-end arrow) with visible cutting edge (triangle).
Histopathology confirmed cellular fibroadenoma. Patient is on follow-up.
Fig. 5 Evolution of hematoma in relation to vacuum-assisted breast biopsy guiding modality
(stereotactic guided vs. ultrasound guided).
Fig. 6 Evolution of hematoma in relation to vacuum-assisted breast biopsy needle gauge (10G
vs. 7G).
Table 2
Results
|
Variable
|
Value
|
95% confidence interval
|
|
Accuracy
|
98.3%
|
91.06–99.96%
|
|
Sensitivity
|
96%
|
79.65–99.90%
|
|
Specificity
|
100%
|
90–100%
|
|
Negative predictive value
|
97%
|
83.7–99.6%
|
Table 3
Lesion characteristics
|
ACR BI-RADS category
|
No. of VABB (%) malignant
|
|
3
|
1 (2)
|
Nil
|
|
4a
|
27 (45)
|
6/27 (22%)
|
|
4b
|
16 (27)
|
7/16 (44%)
|
|
4c
|
11 (18)
|
8/11 (73%)
|
|
5
|
5 (8)
|
4/5 (80%)
|
|
Biopsy histology
|
|
Benign
|
35 (58)
|
|
Malignant
|
25 (42)
|
|
Underestimation (%)
|
4%
|
Abbreviations: ACR BI-RADS, American College of Radiology Breast Imaging Reporting
and Data System; VABB, vacuum-assisted breast biopsy.
Table 4
Pain scores using the visual analog scale
|
Pain score
|
Number of VABB (%)
|
|
Mild (1–3)
|
37 (62)
|
|
Moderate (4–6)
|
21 (35)
|
|
Severe (7–9)
|
2 (3)
|
|
No pain, worst pain experienced
|
0
|
Abbreviation: VABB, vacuum-assisted breast biopsy.
Discussion
The current study affirms the high accuracy of VABB, particularly in small and indeterminate
lesions, with high specificity of 100%, sensitivity of 96%, negative predictive value
(NPV) of 97%, and accuracy of 98%. These results are in line with meta-analysis by
Yu et al[13] estimating specificity of 100% and sensitivity of 98% including larger studies like
those by Penco et al (n = 4,086) (sensitivity = 99.7–100%)[14] and Kettritz et al (n = 2,874) (sensitivity, NPV > 99%).[15]
The DCIS underestimation rate of VABB (4%) in this study was lower than that seen
in previous studies such as Tsai et al (16.7%), Penco et al (17.9%), Cassano et al
(16.7%), and Suh et al (16.1%) and the published underestimation rates of CNB (20–55%),[8]
[14]
[15]
[16]
[21]
[22]
[23] thus reiterating the benefits of VABB over CNB. VABB successfully addressed all
cases of repeat biopsies initially sampled using a 14G CNB automated gun and two cases
of CNB underestimation; hence, our findings are in tandem with existing literature
suggesting lower sampling error and underestimation in VABB.
VABB specimen histopathology revealed 42% lesions as malignant, and 58% as benign,
in concordance with published literature reflecting appropriate patient selection
as per BI-RADS categorization[14]
[15]
[24]
[25]
[26]
[27]; 63% of patients in this study had lesions measuring <10 mm which was comparable
to study cohorts of Kettritz et al (58%) and Penco et al (46%) conforming to appropriate
patient selection for VABB.
No major complications were encountered in this study which is similar to findings
of various previous studies.[28]
[29]
[30]
[31] Like Park and Hong, the most commonly associated adverse event was found to be hematoma
formation, which was manageable by manual compression and with all except one case
having postprocedural hematoma volume <5 mL, which also regressed by the seventh day.[32] Simon et al reported prolonged post-VABB bleeding (>10 minutes) in 7% of patients
and vasovagal response in 1% of the procedures,[33] Johnson et al[34] reported infections requiring intervention in 2% cases and Kettritz et al reported
complications in 1.4% procedures including hematomas >4 cm (n = 25), persistent bleeding (n = 4), vasovagal episodes (n = 5), seizure (n = 1), and inflammation (n = 5).[15] One of the reasons for smaller hematoma volumes observed could be meticulous avoidance
of traversing vessels by Doppler evaluation of skin site entry, trajectory, up to
the lesion, on U-VABB, and repositioning or rolling of breast in case of overlapping
coursing vessels in S-VABB. No statistically significant association between complications
and needle gauge or guiding modality was observed in this study, similar to findings
published by Burbank et al. Alike Bohan et al's experience (55%), most patients in
this study reported mild pain (62%).[35] Mean pain score was 3 which was close to findings of Seely et al (3.1), pointing
to the good overall tolerance of VABB.[36]
Retrospective design of this study, small sample size, and limited follow-up are limitations
of this study. However, VABB holds maximum potential in carefully selected breast
lesions considering the balance between associated cost (higher than CNB) and clinical
impact.
Conclusion
VABB is a promising means of targeting indeterminate or suspicious findings on MG
and ultrasound, and VABB, performed with adequate quality assurance, is safe and efficacious.
Side effects are minimal and hematoma formation is unrelated to gauge of needle and
imaging modality of guidance.
