Image-guided breast biopsy and localisation: recommendations for information to women and referring physicians by the European Society of Breast Imaging

Percutaneous image-guided needle biopsy is essential in the management of suspicious breast lesions detected by screening or during the assessment of clinical abnormalities. It is a safe and cost-effective procedure allowing for an accurate diagnosis, pivotal for adequate decision-making, including, when indicated, treatment planning. Percutaneous image-guided breast biopsies have almost entirely replaced diagnostic surgical excisions that were associated with longer hospital stay, higher cost, and possible complications. In 2010, the European Society of Breast Cancer Specialists, EUSOMA, suggested that 90% of all the women with breast cancer (invasive or ductal carcinoma in situ [DCIS]) should have a preoperative diagnosis by means of percutaneous biopsy [1].

Moreover, the increasing rate of nonpalpable breast lesions detected in screening programmes as well as the general goal of reducing the extent of surgical treatment have increased the need for localisation before surgery. Localisation can also be performed when neoadjuvant therapy is under consideration in order to mark the lesion site for re-evaluation and treatment planning. It is routinely performed using same image guidance techniques used for biopsy and allows for conservative surgical excision of a limited amount of tissue, yielding together an effective treatment and good aesthetic results [2, 3].

Different modalities are available for image-guided breast biopsy and localisation procedures, each of them with their own strengths and weaknesses [4‐6]. The most appropriate method is chosen by radiologists for each individual case. Breast radiologists covering the full spectrum of breast imaging and percutaneous tissue sampling techniques (including the use of markers) and presurgical localisation methods are the most suitable professionals for choosing the optimal technique. When localising methods imply the use of radiotracers or radioactive seeds, radiopharmacy and radio safety training (or cooperation with nuclear medicine/radiotherapy departments) is needed.

This article is the fifth of a series of recommendations for women’s information, all issued by the European Society of Breast Imaging, EUSOBI, the first [7] and the third [8] focusing on mammography, the second on breast magnetic resonance imaging (MRI) [9], and the fourth on breast ultrasound (US) [10]. The current article represents also an update of a previous EUSOBI guideline regarding diagnostic interventional breast procedures, published in 2007 [11]. It is in particular addressed to patients for whom an image-guided breast biopsy or localisation is, or may be, under consideration and to physicians dealing with these patients. In particular, eight special information notes (from A to H) and seven frequently asked questions (FAQs) are formulated for direct communication with women. Considering the differences across European countries in terms of available technology, national guidelines, clinical practices, health care systems, and insurance coverage, the applications of these recommendations can vary under local conditions.

A search on the PubMed/Medline has been performed for papers published from January 2009 to March 2019, using the terms “breast” AND “biopsy” OR “fine needle” OR “localization” OR “marker” OR “interventional”. Articles with an informative content most suitable for the purpose of these recommendations were selected as references with special regard to predetermined issues: safety/quality, protocols and techniques, test performance (sensitivity and specificity), and clinical indications. Other articles were included when found to be important among the references of the retrieved articles or when suggested by one or more authors. The entire text underwent a double evaluation by the authors, each of them contributing with relevant intellectual content. However, as many different topics are considered, single authors generally agreeing on these recommendations may have different opinions on individual statements.

This article summarises the information to be provided to women and referring physicians about percutaneous breast biopsy and presurgical localisation under the guidance of mammography/tomosynthesis, US, and MRI.

Currently, even taking into account recent advances in breast imaging, tissue sampling represents the most accurate method for confirmation or exclusion of malignancy [4]. In fact, there are a variety of benign abnormalities which can mimic malignancy on all breast imaging modalities, i.e., mammography and other x-ray techniques (including tomosynthesis and contrast-enhanced mammography), US, and MRI. The reason to perform a percutaneous biopsy is to prevent unnecessary surgery, associated morbity and costs for equivocal findings on imaging with final non-malignant histopathology. In addition, tailored treatment strategies are currently available including chemotherapy and hormonal therapy before or after surgery (so-called neoadjuvant therapy or adjuvant therapy, respectively), surgical options from lumpectomy to mastectomy with immediate reconstruction, whole or partial breast radiation therapy. The choice among all these options is influenced not only by imaging findings (especially in relation to disease extent) but also by the diagnosis based on percutaneous tissue sampling, in particular when the analysis includes not only basic morphological characteristics but also the molecular pattern of the tumour [11]. Moreover, needle sampling of axillary lymph node, when indicated, adds information for treatment planning [12‐14].

We will describe here the details of the different technical options for breast tissue sampling—fine-needle sampling (FNS), core needle biopsy (CNB), and vacuum-assisted biopsy (VAB)—and the imaging modalities for guidance.

We highlight the results of the latest systematic review of the literature and meta-analysis, published in 2014 by the United States Agency for Healthcare Research and Quality [15]. Based on 160 studies using CNB and VAB techniques, the authors found that both US-guided and mammography-guided biopsies had average sensitivities over 97% and specificities ranging from 92 to 99% while non-imaging-guided free-hand biopsy methods had an average sensitivity of 91% and specificity of 98%. Considering that free-hand non-imaging-guided biopsies are performed only on large palpable lesions, image-guided biopsies had better results for the more demanding task of sampling nonpalpable, smaller lesions. For this reason, free-hand breast biopsy cannot be recommended whenever image-guided biopsy is available. CNB and VAB devices (under the same imaging guidance) had similar performances; CNB was found to be associated with a lower risk of adverse events and complications than open surgical biopsy, which were sparsely reported (e.g., 2–10% haematomas, 4% repeat biopsy, 4–6% infections, 2% abscesses). The incidence of adverse events with CNB was found to be 1–1.5% and that of severe complications less than 1% for all needle sampling techniques; VAB appeared to be associated with increased bleeding and haematoma formation; biopsies performed with patients seated upright appeared to be associated with increased risk of vasovagal reactions; CNB obviated the need for surgery procedures in about 75% of women. The authors concluded that the evidence suggests that US-guided and mammography-guided biopsies have sensitivity and specificity close to that of surgical biopsy with fewer adverse events and that non-imaging-guided free-hand procedures have lower sensitivity than image-guided methods. This large literature review clearly explained the reasons for recommending image-guided biopsy instead of surgical biopsy as a general rule of good practice. However, with regard to VAB versus CNB, lesion size and type must be considered and this may have resulted in an underestimation of the potential advantages of VAB over CNB, particularly when using mammography, tomosynthesis, or MRI guidance.

It is important to note that percutaneous needle biopsy may not provide a definitive diagnosis when the histopathological report describes the presence of a lesion with uncertain malignant potential (also called high-risk or B3 lesion). This occurs in 3 to 9% of cases, with a range of rates turning out to be malignant (10–33% or also higher rates) [10, 16‐25]. These lesions include atypical ductal hyperplasia, benign phyllodes tumours, flat epithelial atypia, classical lobular neoplasia, papillary lesions, radial scars, and other rare entities. Each of them, when surgically removed, shows a variable upgrade rate to invasive cancer or DCIS [26]. Although a small but significant increase of imaging surveillance has been described for classical lobular neoplasia, flat epithelial atypia, and papillary lesions diagnosed on VAB instead of surgical removal, from 24 to 35% of high-risk/B3 lesions (in particular, atypical ductal hyperplasia and phyllodes tumours), are recommended for surgery [27]. In addition, breast specialists should take into account that women diagnosed with high-risk/B3 lesions have a long-term moderately increased risk of breast cancer [28].

Women should be informed that the radiologist may propose a repeat needle sampling or surgical intervention also in the case of biopsy resulting into normal breast tissues or benign abnormalities. After a negative breast tissue needle sampling, imaging follow-up is usually planned, with imaging modalities and time interval to be defined for each individual case.

Note A. Percutaneous image-guided biopsy has replaced surgical biopsy allowing a minimally invasive safe, accurate, and cost-effective diagnosis of breast lesions, necessary for the definition of treatment planning. In the case of a biopsy resulting in a pathological diagnosis of a high-risk/B3 lesion, discuss with your radiologist and the breast care team the best option for you (either repeat biopsy, surgical removal, or imaging surveillance). After a negative image-guided tissue sampling, imaging follow-up is usually planned, with imaging modalities and time interval to be defined for each individual case.

Different percutaneous image-guided techniques are available to diagnose palpable and nonpalpable breast lesions. In the last decades, they have improved patient management, avoiding unnecessary surgical biopsy for benign lesions [29]. Nowadays, FNS, CNB, and VAB coexist, the first providing material for studying cells (cytological examination), the last two providing material for studying tissues (histopathological examination). All these techniques can be theoretically guided by mammography, tomosynthesis, US, or MRI. However, FNS and CNB should be used under US guidance, VAB under mammography/tomosynthesis or MRI guidance. These prevalent combinations are due to both technical considerations, including visibility at each technique and lesion types detected. Breast size, lesion location and size as well as local availability of instrumentation and expertise are taken into account. Table 1 shows the indications for the combinations between imaging guidance and sampling type.

Whenever the lesion is well identified on US, this technique is preferred due to the easy approach and the short duration of the procedure (implying a more comfortable woman’s experience) as well as a lower cost. If a lesion is not clearly identifiable on US, mammography, tomosynthesis or MRI is used for guidance, typically the former in the case of suspicious calcifications and architectural distortions and the latter in the case of lesions only visible on MRI [30].

Needles of different size and length are adopted for percutaneous image-guided biopsies. The diameter is described by gauge numbers. Differently from an intuitive reasoning, smaller gauge numbers indicate larger needle diameters. Commonly applied needles have a diameter ranging from 0.4 (27 gauge) to almost 4.6 mm (7 gauge).

Image guidance outperforms non-imaging-guided free-hand approaches for breast lesion sampling. The choice among mammography/tomosynthesis, US, or MRI methods is made by the radiologist according to several factors, the most important being the visibility of the lesion to be targeted.

Axillary lymph node assessment is an integral part of preoperative staging in patients with newly diagnosed invasive breast cancer [12]. Information about possible metastatic involvement of axillary lymph nodes can help avoid unnecessary procedures, triaging patients directly to axillary lymph node dissection [13]. Axillary lymph node US is the easiest way to identify abnormal lymph nodes. Tissue sampling with FNS or CNB is similar as in the breast. CNB seems to be more accurate than FNS in the diagnosis of axillary lymph node metastasis [14]. The incidence of adverse events is rather low; however, complications can be more severe than those of breast tissue sampling. After a biopsy, a clip may be placed in the biopsied lymph node, and marking of lymph nodes with clips and other devices is currently under investigation.

For any combination of guidance methods and needle types, concordance between the cytological or pathological result and the radiological appearance of the targeted lesion has to be verified. This mandatory task must be performed by the professional who performed the image-guided sampling, usually a breast radiologist, in collaboration with the reporting pathologist [75‐78].

In most cases, the radiologist will consider the cytology or pathology result as concordant with imaging findings, especially when CNB or VAB has been performed. However, in the case of inadequate sampling (more frequent in the case of FNS) or discordance between cytology/pathology and imaging findings, a repeat biopsy using the same or a different biopsy method can be considered, which is often after discussion in the multidisciplinary meeting. In the case of findings highly suspicious for malignancy on US, mammography/tomosynthesis, or MRI with discordant benign findings at FNS/CNB/VAB, it is also possible to go directly for surgical removal, after preoperative image-guided localisation, if the lesion is nonpalpable.

Percutaneous breast biopsies are minimally invasive and in general very safe, with severe complications requiring treatment being exceedingly rare [15, 34]. Common side effects include minor pain and bruising [41]; other complications are large haematomas or infections, at a rate of one every 1000 procedures [77, 79].

The risk of bleeding will increase with the needle diameter and the amount of sampled tissue [15]. Bleeding after biopsy is usually self-limited or limited by manual compression for 5–10 min (the time may be longer in the case of arterial bleeding) but may produce mild discomfort and pain for several days. Severe bleeding requiring surgical intervention is very rare. Careful screening for bleeding disorders, avoidance of vessels when choosing the needle track (possible only under US guidance using colour-Doppler imaging), and adequate compression after the procedure reduce the probability of severe bleeding.

Pseudoaneurysm (also called false aneurysm) of breast vessels is a very rare complication after either CNB or VAB, reported in the literature mainly as case reports. It is due to blood leaking from a breach in the arterial wall, contained by the adventitia or surrounding perivascular soft tissue. An enlarging mass a few days after the procedure, showing blood flow inside, is the commonest presentation. Management options include observation, US-guided focused compression, thrombin injection, open surgical repair, and percutaneous embolisation [80].

As with any percutaneous intervention, a low risk of infection does exist, so using sterile packed devices and keeping the environment clean as much as possible is important. The risk of infection after breast biopsies is extremely rare, reported to be 0.1% for US-guided biopsies [81]. They are mostly limited to skin or soft tissue and highly responsive to oral antibiotics. The risk of infection may be higher in patients with diabetes or a compromised immune system.

The risk of mechanical displacement of malignant cells along the biopsy tract can very rarely occur and is referred to as neoplastic seeding. The incidence of this event was reported to be 2 of 1644 biopsies [82] or, more recently, 8 of 4010 biopsies, [83], meaning that one or two of such events are expected every one thousand biopsies, likely because the displaced tumour cells are usually not viable [84]. Their biological significance is disputed, since such seeding is mostly located in the skin and detected early and thus removed without consequences. A retrospective analysis [85] including 719 patients after conserving surgery and radiotherapy for stage I and II breast cancer (189 patients having preoperative CNB and 530 without, with a median follow-up of 78 and 71 months, respectively) did not show increase of the local recurrence rate for the CNB group.

A study [86] reported an increase in distant metastases at long-term follow-up for a cohort of patients who underwent CNB versus FNS. However, the study has relevant limitations, the most important being the retrospective design, the small sample size and the times of cancer diagnosis of the two cohorts (203 patients diagnosed from 1991 to 1995 for CNB; 181 patients diagnosed from 1971 to 1976 for FNS). The authors adjusted for the difference in treatment over time. However, the 20-year time difference implies a greater sensitivity for metastases in the CNB cohort as a non-negligible source of bias [87]. In addition, no prospective studies replicated these findings.

A severe, but extremely rare complication of a free-hand or US-guided breast biopsy is the development of a chest wall injury, for example a pneumothorax, especially if an inexperienced examiner uses an improperly steep angle for access [88]. A pneumothorax is an abnormal collection of air between the lung and the chest wall and presents with sudden onset of sharp, one-sided chest pain and shortness of breath. The risk is practically null with mammography or MRI-guided biopsy, because the needle is introduced parallel to the chest wall.

Note D. Percutaneous breast biopsy is a safe procedure, with an extremely low risk of serious complications needing surgery or emergency assistance. Mild complications are also very rare: a certain degree of bleeding is the most common side-effect, usually self-limited; skin or soft-breast tissue infections respond well to oral antibiotics. The probability of neoplastic seeding along the biopsy tract is 2 of 1000 biopsies or lower and considered irrelevant due to the adjuvant radiation and pharmacological treatment in case of breast cancer. All mentioned complications are less frequent than complications under surgical biopsy with or without full anaesthesia.

The planned procedure, including the rationale for performing the biopsy/localisation, potential benefits, possible complications, and likely outcomes must be explained in detail to the patient and informed consent should be obtained in advance (e.g., when planning the procedure or 1−2 days before the procedure) or immediately before the procedure [51, 58].

The patient should be informed about the reasons for performing a needle biopsy or a localisation procedure.

Before the procedure, informed consent must be obtained according to national/local regulations.¹ Information must be given on the devices (needles) used for sampling and for guiding the sample, on the frequency of adverse events, side effects, and complications, better if using natural frequencies instead of percentages or complex epidemiological indices, including the need for repeat biopsy or surgical excision in cases of inadequate sampling, poor radiological-pathological concordance, or pathological diagnosis of high-risk/B3 lesions, as explained above.

Patients should be informed about the possible necessity of placing a marker clip in the biopsy site, taking into account that some women may have objections against potentially permanent placed foreign bodies in their breasts. In those cases, precise information on the risks associated with not using these markers has to be provided to the patient, especially the risk of missing a small lesion requiring surgery that may not be visible after the biopsy.

Efforts should be made to reduce the patient’s anxiety prior to the procedure, as the anticipated pain strongly correlates with the level of pain experienced by the patient during the procedure, so that communication with patients before biopsy regarding minimal average pain reported during biopsy is a good strategy [91]. The quality of human communication between the radiologist and the patient has been demonstrated to impact on patient’s anxiety [92]. A study reported that listening to guided meditation lowered biopsy pain during biopsy and that meditation and music reduced patient anxiety and fatigue without a negative impact on radiologist-patient communication [93]. We highlight that a good organisation of the medical team, a calm and efficient workflow, and a constant communication with the patient before, during, and after the procedure ensure a high compliance.

The experience of a patient needing a biopsy differs according to the image guidance adopted, as explained in the previous sections. While US-guided is in most cases a straightforward and fast procedure lasting no more than 15 min, mammography/tomosynthesis- and MRI-guided biopsies require preparatory technical measures aiming at the exact lesion localisation, leading to procedural times that can exceed 30 min and increase patient anxiety. Some evidence exists that tomosynthesis guidance can reduce the procedural time in comparison with stereotactic guidance [64]; one study [65] comparing 706 procedures under tomosynthesis guidance versus 439 procedures under stereotactic guidance showed an over 50% reduction in procedural time and about 75% reduction in radiation exposure.

Whichever combination of guidance and needle is used, once the patient is positioned and the best and safest needle access route to the lesion has been chosen, the corresponding skin is cleaned and disinfected. Whereas for FNS the use of local anaesthesia is optional (the size of the needle used is similar to an anaesthesia needle and the anaesthetic liquid could interfere with cytological aspiration), all other types of percutaneous breast biopsy are usually performed under local anaesthesia.

For superficial anaesthesia, lidocaine buffered in sodium bicarbonate may be used to reduce the initial stinging sensation of the lidocaine injection [41, 58]. In the case of allergy to lidocaine, alternative drugs have to be used. Even with optimal local anaesthesia, some discomfort or pain may be felt during needle insertion and tissue sampling, which will vary significantly from patient to patient [91].

At the completion of the biopsy, local manual compression, usually for 5 min, as well as cooling, may be applied to the biopsy site to achieve haemostasis and to minimise the amount of bleeding. In addition, the application of a compression bandage can be applied, in particular in the case of moderate or severe bleeding,

Note F. Patients’ experience will be different depending on the type of device used for sampling and the imaging guidance adopted. A percutaneous biopsy may last from 15 to 45 min and is generally well tolerated. The position will be the same as the diagnostic examination for US (supine) or MRI-guided biopsy (prone) and may be prone, seated, or in lateral decubitus for mammography/tomosynthesis-guided biopsy, according to the adopted system. Communication between the radiologist and the patient is crucial for anxiety reduction.

Some centres have developed rapid diagnosis as “one-stop breast clinics”, providing the results of the biopsy immediately after the procedure. This type of management has been proven to be cost-effective and accurate [94]. However, efforts should be made to decrease distress and anxiety of the patients also in this setting in regards to the quality of the patient-doctor relationship and doctors’ interpersonal skills, doctors’ availability, and waiting time [95], which are important general aspects to be improved in breast image-guided interventions.

Note G. At the end of biopsy, the patient is observed in the radiology department for 15–60 min. She can then be discharged and adequate instructions on common side effects and complications, as well as corresponding remedies are suggested. The next visit is planned for communication and discussion of the results, including the radiologist who performed the procedure and participated in the assessment of concordance between cytology/pathology diagnosis and lesion appearance at breast imaging.

Nonpalpable, clinically occult breast lesions amenable to surgery need image-guided preoperative localisation. This allows guidance of surgeons to the lesion for a safe and effective intervention aiming at obtaining both a complete removal (clear margins) and a good cosmetic result [2, 3, 96]. Radiologists play a crucial role in this.

Methods for preoperative localisation of nonpalpable breast lesions are carbon marking, wire localisation, radiotracer injection (usually called radio-guided occult lesion localisation, ROLL), radioactive seed localisation, and magnetic seed localisation (Table 2).

Image-guided percutaneous needle biopsy represents a fundamental technique to characterise the nature of suspicious breast abnormalities and has almost completely replaced the surgical biopsy for both palpable and nonpalpable lesions. Radiologists play a crucial role in the detection as well as diagnosis and management of breast disease. Depending on the patient profile and lesion characteristics, the radiologist will choose the best available biopsy system from FNS, CNB, or VAB. The choice will also depend on local experience and availability. When a lesion is visible on US, this is the preferred method for biopsy being most accessible, comfortable, and straightforward compared to other techniques. Minor pain and bleeding are possible post-biopsy sequelae, but the risk of severe complications is very low. Radiologic-pathologic correlation is essential for an accurate and successful conclusion of the diagnostic procedure.

Preoperative localisation of nonpalpable lesions amenable to surgery is available through US, mammography/tomosynthesis, or MRI guidance, typically by wire insertion.

Needle biopsies are percutaneous interventional procedures performed with a needle under the guidance of a breast imaging modality; depending on the needle size, local anaesthesia is administered. Excisional biopsy is a surgical operation aiming at removing the entire lesion to be analysed by the pathologist; it may be performed under local or general anaesthesia.

FNS is percutaneous lesion sampling, usually performed under US guidance, using the same fine needles suitable for intramuscular injections or thinner needles, according to local experience. Once the needle is in the lesion, multiple cells are collected; the biological material is prepared as smears on glass slides and examined by the pathologist for a cytological diagnosis. CNB is a percutaneous procedure that uses a spring-loaded needle for “true cutting” tissue “cores” providing a histological tissue diagnosis. VAB performs a task similar to that of CNB using larger needles, facilitated by a vacuum-generating device that attracts the tissue towards the needle, allowing removal of larger amounts of tissue in comparison to CNB.

Minor pain and bruising can occur immediately at the end of the procedure and up to one or two days. If necessary, adequate analgesic support can be taken.

Yes, local anaesthesia is always administered before core needle biopsy (CNB) or vacuum-assisted biopsy (VAB). Before FNS, as the procedure for delivering anaesthesia is not different from collecting cells, local anaesthesia is optional.

The risk of mechanical displacement of malignant cells along the biopsy tract can rarely occur and is referred to as “neoplastic seeding”. It has been reported to happen in less than 2 cases every 1000 biopsies. It is considered oncologically irrelevant due the high probability of lack of viability of the displaced cells and to the possible addition of adjuvant radiation and pharmacological treatment in the case of breast cancer. In fact, preoperative needle biopsy does not increase local recurrence rate in breast cancer patients. In addition, if seeding does occur, it is mostly seen in the skin, which is detected early and easily treated, with low clinical importance.

A very low risk of infection accounting for 1 case every 1000 procedures does exist, and adherence to sterile working conditions is important. Infectious complications are limited to skin or soft tissue and respond well to oral antibiotics. The risk of infection may be higher in patients with diabetes or compromised immune systems.