Ann R Coll Surg Engl 1998; 80: 253-256
Ultrasound-guided Tru-cut biopsy of the
N P Woodcock MB ChB
Senior House Officer in Surgery
I Glaves FRCR
D R Morgan MRCPath
John MacFie MD FRCS
Scarborough Hospital, Scarborough
Key words: Ultrasound; Tru-cut needle biopsy; Breast lesions
biopsy may be used as an alternative to fine needle
aspiration cytology for the assessment of discrete
mass lesions of the breast. This is a retrospective
study of 187 biopsies, comparing the results with a
final diagnosis obtained from subsequent excision or
outpatient follow-up. Biopsies were performed using
a spring-loaded gun under ultrasound guidance.
biopsies, 98 of which were subjected to surgery, with
no false-positives. Twelve biopsies contained 'atypical
cells', pre-invasive malignancy or risk factors for
invasive carcinoma, ten ofwhich proved to be invasive
malignancy on excision. Normal or benign tissue was
found in 61 biopsies, but of those that proceeded to
excision biopsy, 16 were invasive or in situ carcinoma.
The sensitivity of the procedure for detecting sig-
nificant pathology was 88.7%, and the specificity 100%.
When used as part oftriple assessment, the sensitivity
increases to 97.9%.
well-tolerated and reliable procedure for providing a
treatment, and obviating the need for formal exci-
sion biopsy of lesions for which there is a low index of
Accurate preoperative diagnosis
management of patients with palpable breast lesions. It
is well established that the best means of achieving this is
by triple assessment comprising clinical examination,
breast using mammography and/or
ultrasonography, and direct examination of cells or tissue
from the lesion itself. In most centres this latter is
(FNAC), which is now considered by many authorities
to be an essential prerequisite for any breast clinic (1).
However, FNAC does have disadvantages; optimal
results necessitate the availability of
committed cytopathologist and the technique demands
experience and expertise if the incidence of inadequate
samples is to be minimised (2,3). Provisional work in our
own unit showed FNAC
inadequate sample rate of >30%. Most hospitals do not
have a dedicated cytology service and, theoretically, this
should preclude them from the management of patients
with symptomatic breast disease.
automated Tru-cuts biopsy device under ultrasound
guidance for abdominal mass lesions demonstrated that
this technique was associated with low morbidity, little
pain and the facility to provide our pathologists with
tissue specimens that were amenable to conventional
histological techniques (4). This, together with other
reported experience of Tru-cut biopsies in the manage-
ment of patients with palpable breast lesions, prompted
our adoption of this technique as part of our standard
triple assessment of patients with symptomatic breast
disease in place of FNAC.
a trained and
to be associated with
Patients and methods
This is a retrospective study looking at ultrasound-guided
automated Tru-cut needle biopsies (TCNB) of breast
lesions performed between April 1994 and June 1996. All
the patients were under the care of one surgeon and
attended a dedicated breast clinic. Clinical findings were
Surgeon, Scarborough Hospital, Scarborough, North Yorkshire
to: Mr J MacFie MD FRCS, Consultant
N P Woodcock et al.
recorded as normal, benign, suspicious or malignant.
Breast imaging either by mammography or ultrasound
was usually performed
attendance, with the TCNB performed at this time in
the majority of cases. Radiological findings were also
classified as benign, suspicious or malignant, and one
radiologist was responsible for reporting.
Tru-cut needle biopsy was performed under local
anaesthesia by one consultant radiologist using a 14G or
18G needle mounted in
employing a spring-loaded mechanism (Manan Promag
1.2, Manan Medical Products Inc., Northbrook IL60062,
USA). This can be operated with one hand, facilitating
simultaneous ultrasound scanning, performed using a 7.5
MHz linear array real-time probe. The biopsy device
automatically advances the cutting needle into the biopsy
site. The needle tip appears as a bright echo, and its track
during biopsy is demonstrated to confirm that the correct
area has been sampled. Usually three passes were made in
The biopsies were fixed in 10% formalin and embedded
in paraffin wax, and stained with haematoxylin and eosin
for histological examination. Additional immunohisto-
chemistry was performed on selected samples if felt to be
The histology obtained on TCNB was compared with a
final diagnosis obtained from subsequent excision biopsy,
wide local excision or mastectomy. In a minority of cases
no definitive surgery was undertaken, and the patient
retumed to outpatient follow-up.
A total of 187 Tru-cut needle biopsies was performed in
183 patients. The age range of the patients was 29 to 87
years, mean age 61.2 years. Of the four patients in whom
metachronous lesions, while the other two underwent
repeat biopsy of the same lesion. The findings on clinical
and radiological examination are shown in Table I.
Invasive carcinoma was found in 113 TCNBs. Ofthese,
97 patients underwent definitive surgery, either wide local
excision or mastectomy. In 96 cases, the diagnosis of in-
vasive breast carcinoma was confirmed, the other being a
cutaneous metastasis from small cell bronchial carcinoma
on excision biopsy. One patient had a non-Hodgkin's
lymphoma on TCNB and proceeded
because of its large size, without axillary dissection.
two had biopsies of
Surgery was not undertaken in 16 elderly patients who
were managed solely with adjuvant hormonal therapy.
Four of the TCNBs were reported as showing 'atypical
cells', with repeat TCNB or excision biopsy recom-
phylloides tumour in the fourth. Eight biopsies con-
tained pre-invasive malignancy or risk factors for invasive
carcinoma (ductal carcinoma in situ (DCIS) in four,
lobular carcinoma in situ (LCIS) in one and atypical
ductal hyperplasia in three). Of these patients,
proceeded to surgery, all of whom were found to have
In 61 patients, TCNB demonstrated normal tissue or
various benign changes including
fibroadenoma, duct ectasia and fat necrosis. Eight of
these benign biopsies contained evidence of simple
hyperplasia. Of these
patients, 29 did
surgery, being reassured and returned to follow-up in
the outpatient clinic. All of these patients had lesions for
which there was a low degree of suspicion both clinically
and radiographically. In the remaining 32 patients, there
was either a continued clinical impression of malignancy
despite the negative biopsy (n = 21), persistent abnorm-
ality on subsequent review in the clinic (n = 5) or simply
patient anxiety (n = 6). Each of these patients had an open
surgical biopsy, and 16 were found to have malignancy
(one DCIS), none of which were among the anxious
If malignancy (invasive and pre-invasive) or atypical
malignancy is 86.5%. If LCIS and atypical hyperplasia
are included, this figure rises to 88.7%. There were 16
false-negative biopsies; of these, 15 were subsequently
carcinoma in situ. For 13 of these lesions, there was a
high index of suspicion for malignancy either clinically or
on imaging. Thus, when used
assessment of patients with a breast lump, the sensistivity
for the detection of significant pathology is 97.9%. There
were no false-positive biopsies identified in this series,
giving a specificity of 100%. We appreciate that the
diagnosis ofmalignancy is not confirmed in the 16 elderly
patients not subjected to surgery, but in the absence of
any false-positives among the others, these biopsies can be
assumed to be true-positives.
The size of tumour on excision was recorded in 120 out
as invasive carcinomas and
as part of the triple
Table I. Findings on clinical and radiological examination
Ultrasound-guided breast biopsy 255
of 125 cases of invasive or in situ carcinoma. The range
was 0.5 to 10 cm in diameter, with 60 (50%) being 2 cm
or less, and a further 52 (43.3%) no more than 5 cm in
Diagnostic immunohistochemistry was performed on
six of the Tru-cut biopsy specimens. Three of these were
found to be invasive lobular carcinomas, two were
invasive carcinomas of unspecified type, and the sixth
was a non-Hodgkin's lymphoma.
The results of this retrospective study suggest that TCNB
using an automated firing gun under ultrasound control in
conjunction with clinical and radiological assessment
provides an accurate, reliable and safe means of establish-
ing the diagnosis of patients with palpable breast lesions.
Many previous studies have reported on needle core
biopsies of breast lesions. Some studies have reported
false-negative rates as low as 1-6% (3), but in others the
sensitivities have been rather lower than this, between
71% and 80%
(2,5,6). Although TCNB has been
advocated in those patients who have had an adequate
initial fine needle aspiration performed (7), it has not
been widely adopted in the routine investigation of
patients with breast disease. There are many reasons for
this; in particular, TCNB techniques are considered by
many to be associated with significant morbidity in terms
of pain or haematoma formation, and to be inaccurate,
particularly when compared with FNAC. There were no
causes of significant morbidity in any of the patients in
In the majority of centres in the United Kingdom, fine
needle aspiration cytology remains the preferred method
by which cells are obtained from mass lesions ofthe breast
as part of triple assessment, as recommended by the
National Surgical Co-ordination Group in Breast Cancer
Oncology. It is a straightforward technique, and is easily
performed on a patient's first visit to the outpatient clinic.
However, the procedure is not without its drawbacks.
Inadequate samples may be obtained in between 6% and
47% of cases and false-negatives in 1% to 31% of
patients, with the occasional false-positive result (< 1%)
(8). The cytological specimens
may be missed (9), and differentiating invasive and in
carcinomas is problematical. It is also difficult to
make a definite benign diagnosis on the basis of a FNAC
sample. Most pathologists would concur with the view
that a committed breast cytopathologist is required to
examine accurately and report on samples (9). This raises
the possibility that reliance upon FNAC in those hospitals
without established cytological expertise may be in-
appropriate, and there does appear to be a gradual move
towards the routine use of TCNB as part of triple
In our study, the sensitivity ofTCNB for the detection
of malignant pathology was 88.7%. This compares
can be difficult
favourably with FNAC. Simultaneous localisation with
ultrasound is an essential part of this technique.
the number of passes
radiologist is able to visualise the needle track both
placement of the needle. The use of the automated gun
with a high cutting speed reduces the time for which the
needle is present within the patient. These factors must
contribute to the low morbidity associated with this
technique (10), as was the case in this series. Early studies
that showed a significant morbidity associated with this
technique did not use ultrasound guidance.
The size ofbiopsy needle used in this study was 14G or
18G, the narrower bore 18G being used for small lesions
to minimise mass displacement during biopsy. Other
studies have advocated the use of the larger 14G for all
cases, stating that it consistently produces higher quality,
more intact cores, improving diagnostic accuracy, without
an increase in morbidity (8). However, others claim that
the larger needles do cause significantly more bleeding
The singular advantage of core biopsies is the provision
of tissue to the pathologist that is suitable for conventional
histological techniques. This enables a more precise
pathological diagnosis which facilitates patient manage-
ment. Our results confirm this benefit. For example, we
found that TCNB was very effective in detecting lobular
carcinomas, which FNAC is liable to miss (11). All seven
were correctly identified as invasive carcinoma, three by
using immunohistochemistry. This can be performed
retrospectively on TCNB samples, but with FNAC
multiple smears are necessary in advance just in case
this technique is required.
In each ofthe four patients in whom the Tru-cut biopsy
was reported as showing atypical cells, the final histology
was subsequently found to be invasive malignancy. It
seems clear that patients with these findings on needle
biopsy require further evaluation, ideally consisting of
formal excision biopsy. TCNB is able to confirm the
presence of invasion in a carcinoma, but is less effective at
reliably indicating its absence when only ductal carcinoma
in situ is found (12).
This fact is supported by our
findings, as three out of the four patients with DCIS on
Tru-cut who underwent excision biopsy were each found
to have invasive carcinoma. The single patient with LCIS
on biopsy also had an invasive carcinoma on excision.
The finding of atypical ductal hyperplasia (ADH) on
Tru-cut should be treated with caution, as a low grade
carcinoma cannot be excluded (13). In this study, two of
the three patients with ADH on biopsy were subsequently
found to have an invasive carcinoma. This category of
patient requires excision biopsy (13), repeat TCNB or at
least close observation. A further interesting point is the
fact that four out of the eight Tru-cut biopsies showing
evidence of simple hyperplasia without atypia were false-
negatives, with subsequent discovery of invasive malig-
nancy. It would appear sensible to follow these patients
more closely than those with other benign findings.
Benign changes or normal breast tissue on TCNB need
to be interpreted in the context of the clinical and
N P Woodcock et al.
radiological assessment of the lesion (3). If these are both
consistent with the biopsy findings, the patients can be
reassured as to the absence ofmalignancy and returned to
outpatient follow-up (14). This was the case in 29 of the
patients in our study. In this group, in whom the need for
surgical excision biopsy is removed, the cost of diagnosis
is reduced by more than 50% (15). This has important
considering the increasing numbers of
screening. There is also no unnecessary scarring of the
making it easier to assess satisfactorily at
subsequent examination (6).
whom there is no such correlation between the TCNB
findings and clinical/imaging assessment, excision biopsy
is mandatory if malignancy is to be excluded.
In this series there were 16 false-negatives.
interesting to note that 14 of the cases were subsequently
shown to have tumours of 2.5 cm or less. However, the
other two false-negatives were larger than 5 cm in
diameter. Both appeared to have a necrotic core on
ultrasound, and inadvertent biopsy of this area would be a
possible explanation for the diagnosis being missed in
these cases. It is also noteworthy that not all mammo-
graphically identified breast lesions or even those with
obvious clinical abnormality will have easily definable
abnormality on sonography and this represents a distinct
disadvantage of this technique. This again emphasises the
importance of triple assessment in the management of
patients with breast masses, in that dependence on one or
other diagnostic modality will inevitably be associated
with a significant incidence of false-negative results.
We propose that ultrasound-guided automated Tru-cut
needle biopsy is an accurate and reliable alternative to fine
needle aspiration cytology as part of the triple assessment
ofmass lesions in the breast. It is particularly useful in the
absence of a specialist breast cytopathology service. It is a
relatively straightforward procedure, without significant
morbidity. It allows reassurance of patients with a benign
biopsy in whom clinical and radiological appearances are
of low suspicion, obviating the need for excision biopsy.
When invasive carcinoma is demonstrated, it allows one-
step definitive surgery. However, excision biopsy does
remain necessary if the TCNB shows atypical cells, pre-
malignancy, if a benign biopsy does not correlate with
the clinical impression and/or radiological appearance, or
if a patient remains anxious despite reassurance.
However, in patients in
risk factors for
1 Guidelines for surgeons in the management of symptomatic
breast disease in the United Kingdom Prepared by the
National Surgical Co-ordination Group in Breast Cancer
2 Litherland JC, Evans AJ, Wilson ARM et al. The impact of
core biopsy on pre-operative diagnosis rate of screen-
detected breast cancers. Clin Radiol 1996; 51: 562-5.
3 Dahlstrom JE, Jain S, Sutton T, Sutton S. Diagnostic
accuracy of stereotactic core biopsy in a mammographic
breast cancer screening programme. Histopathology 1996; 28:
4 Jaeger HJ, MacFie J, Mitchell CJ et al. Diagnosis of
abdominal masses with percutaneous biopsy guided by
ultrasound. BMJ 1990; 301: 1188-91.
5 Dowlatshahi K, Yaremko ML, Kluskens LF et al.
palpable breast lesions: findings of stereotaxic needle-core
biopsy and fine needle aspiration cytology. Radiology 1991;
6 Chare MJB, Flowers CI, O'Brien CJ, Dawson A. Image-
guided core biopsy in patients with breast disease. Br 7 Surg
1996; 83: 1415-16.
7 Carty NJ, Ravichandran D, Carter C et al. Randomised
comparison of fine-needle aspiration cytology and Biopty-
Cut needle biopsy after unsatisfactory initial cytology of
discrete breast lesions. Br J Surg 1994; 81: 1313-14.
8 Parker SH, Lovin JD, Jobe WE et al.
biopsy with a biopsy gun. Radiology 1990; 176: 741-7.
9 Rode J. Commentary: fine needle cytology versus histology.
Histopathology 1989; 15: 435-9.
10 Parker SH, Jobe WE, Dennis MA et al.
automated large-core breast biopsy. Radiology 1993; 187:
11 Sadler GP, McGee S, Dallimore NS et al. Role offine-needle
aspiration cytology and needle-core biopsy in the diagnosis of
lobular carcinoma of the breast.BrJISurg 1994; 81: 1315-17.
12 Liberman L, Dershaw DD, Rosen PP et al. Stereotaxic core
biopsy of breast carcinoma: accuracy at predicting invasion.
Radiology 1995; 194: 379-81.
13 Jackman RJ, Nowels KW, Shepard MJ et al. Stereotaxic
large-core needle biopsy of 450 non-palpable breast lesions
with surgical correlation in lesions with cancer or atypical
hyperplasia. Radiology 1994; 193: 91-95.
14 Mikhail RA, Nathan RC, Weiss M et al. Stereotaxic core
needle biopsy of mammographic breast lesions as a viable
alternative to surgical biopsy. Ann Surg Oncol 1994; 1: 363-7.
15 Liberman L, Fahs MC, Dershaw DD et al. Impact of
stereotaxic core breast biopsy on cost of diagnosis. Radiology
1995; 195: 633-7.
Received 25 February 1998