ORIGINAL ARTICLE

ACTA RADIOLOGICA

Large-Core Needle Biopsy versus Fine-Needle Aspiration Biopsy in Solid Breast Lesions: Comparison of Costs and Diagnostic Value S.-M. VIMPELI, I. SAARENMAA, H. HUHTALA & S. SOIMAKALLIO Medical Imaging Centre, Department of Radiology, Pirkanmaa Hospital District, Tampere, Finland Department of Radiology, Hatanpa¨a¨ City Hospital, Tampere, Finland Tampere School of Public Health, Tampere University, Tampere, Finland Vimpeli S-M, Saarenmaa I, Huhtala H, Soimakallio S. Large-core needle biopsy versus fine-needle aspiration biopsy in solid breast lesions: comparison of costs and diagnostic value. Acta Radiol 2008;49:863
869. Background: In the current climate of budget constraints and personnel shortages, hospitals are required to demonstrate ever-greater cost effectiveness. In the diagnosis of breast lesions, it is useful to compare the costs and benefits of fine-needle aspiration biopsy (FNAB) and core-needle biopsy (CNB). Purpose: To determine the total costs of FNAB and CNB of breast lesions, and to compare the costs and diagnostic value of these methods. Material and Methods: The material consisted of 688 breast lesion cases from a Finnish breast-imaging unit. All cases underwent FNAB and/or CNB. Based on the primary biopsy method used, the cases were allocated to the FNAB (n 590) or CNB (n 98) group. Data on costs, sensitivity, and specificity were used to compare the cost effectiveness of the two methods. Results: 590 FNABs were performed in the FNAB group and 98 CNBs in the CNB group. In addition, we needed 78 CNBs in the FNAB group and 11 FNABs in the CNB group for further assessment. The false-negative rate for FNAB was 19% and for CNB 11%; the false-positive rates were 9% and 1%, respectively. The average cost of biopsies per case in the FNAB group was t66 and in the CNB group t221. FNAB was most expensive when performed under ultrasound guidance (t44.65) and CNB when performed under stereotactic guidance (t246.37). Conclusion: FNAB is a less reliable and less informative diagnostic method than CNB. Although a negative or indeterminate FNAB result requires follow-up or a re-biopsy with core needle, it is still a cost-effective procedure. Stereotactic guidance considerably increases the costs of CNB, and therefore US guidance should be used whenever possible. Key words: Biopsy; breast; cost effectiveness; mammography Sanna-Mari Vimpeli, Medical Imaging Centre, Department of Radiology, Pirkanmaa Hospital District, PO Box 2000, 33521 Tampere, Finland (tel. 358 3 311611, fax. 358 3 31165586, e-mail. [email protected]) Accepted for publication May 16, 2008

Mass mammography screening programs and advanced imaging techniques, including ultrasonography (US) and magnetic resonance imaging (MRI), have led to a sharp increase in the detection of breast abnormalities. Despite its high costs and invasiveness, surgical biopsy remains the gold standard for the accurate diagnosis of these lesions. However, advances in image-guided biopsy techniques have paved the way for non-surgical diagnosis, which is more cost effective and less traumatic than

surgical biopsy and allows for the planning of onestep surgery (1
9). The benefits of preoperative diagnosis are further underscored by the fact that 80% of all lesions detected by mammography are benign (10). A widely used, noninvasive diagnostic technique is fine-needle aspiration biopsy (FNAB). FNAB is safe, cost effective, easy to perform, and requires hardly any recovery time. It is also highly accurate in determining breast lesion cell type. However, it

DOI 10.1080/02841850802235751 # 2008 Informa UK Ltd. (Informa Healthcare, Taylor & Francis AS)

864

S.-M. Vimpeli et al.

has two main limitations, viz. insufficient sampling rates and its inability to diagnose tumor invasion (11
13). Core-needle biopsy (CNB) is a well-accepted sampling method. It has proved to be superior to FNAB in some cases (e.g., microcalcifications) and is indicated when the FNAB findings are equivocal. CNB provides histological analysis and tumor grading, but it is also more time consuming and expensive than FNAB. Large-core specimens are usually acquired by removing varying volumes of tissue with automated needle guns or vacuumassisted biopsy probes. A vacuum-assisted biopsy probe is more reliable than an automated biopsy gun in lesions that are small or superficial, in thin breasts, or in lesions containing calcifications. The cost of a vacuum-assisted biopsy device is also relatively high. The false-negative rate with CNB is around 2
9%. Most studies have reported a higher false-negative rate for FNAB, with figures range from 2% to 22%, and in some analyses up to over 30% (7, 11, 14
20). Guidance is usually provided by stereotaxis, ultrasound, or palpation, and more recently by MRI. Medical professionals are concerned first and foremost with the accuracy of the biopsy method, but healthcare administrators must also consider the financial implications. It is therefore of considerable interest to know which of these two biopsy methods is the most cost effective. In this study, we assess the overall costs of FNAB and CNB; compare their cost effectiveness; and determine their specificity and sensitivity to see if they are equally reliable.

Material and Methods Study population and biopsy The data were collected in 2002
2004 at a highvolume breast-imaging unit in Finland. The study population consisted of both symptomatic and screening patients who had FNAB and/or CNB of a solid breast lump and who consented to participate (about 98%). Symptomatic patients were referred to the unit by a physician, but some were examined at their own request. Cysts, abscesses, and microcalcifications as solitary findings were excluded, as were palpable lumps not visible with mammography or US. Consent was obtained after the intervention to ensure that the willingness of patients to participate had no effect on the diagnosis. The study was approved by the Tampere University Hospital Ethics Committee. Acta Radiol 2008 (8)

The total number of cases in our study was 688: 590 cases in the FNAB group and 98 in the CNB group. If both types of biopsy were taken, the case was classified according to the biopsy that was taken first. Dual biopsies were usually required to confirm an unspecific FNAB result by CNB, or more rarely the other way round. If both FNAB and CNB were taken on the same day, the case was classified on the basis of the radiologically determined primary biopsy method. The patients ranged in age from 14 to 93 years (median 55 years). One hundred eighty-three (27%) patients came to the unit without a referral; 505 (73%) had referrals from GPs or specialists. The most frequent symptom was a lump, which was present in 349 (51%) patients. Suspected lumps were examined by palpation, mammography, and US. Findings were classified into three categories, viz. benign, malignant, and suspicious for malignancy. Mammographic findings were classified into modified BI-RADS categories: 1 normal, 2  benign, 3low suspicion of malignancy, 4 high suspicion of malignancy, 5malignant (21). To confirm the benign or malignant nature of a lesion, FNAB or CNB or sometimes both were performed under guidance of US, stereotaxy, palpation, or various combinations. FNAB was preferred for all mass lesions visible on US. CNB was used primarily when the lesion was only visible on mammography, or when FNAB had produced an unspecific result. The choice of guidance method was based on a number of factors, including the visibility of the lesion by different methods, the availability of equipment, or simply the radiologist’s preference. MRI was not available. FNAB was performed under guidance of US or palpation or both. If necessary, sonographic equipment (Diasus; Dynamic Imaging, Livingston, UK) was used using an 8
16 MHz/5
10 MHz linear transducer. The biopsy was obtained from the most representative part of the tumor using a free-hand technique. CNB was performed with a 14G (in some cases 16G) spring-loaded needle under guidance of stereotaxy, US, palpation, or USpalpation. In US guidance, the needle was set in the middle of the tumor using a free-hand technique. Most typically, 4
5 specimens were obtained in various directions (four or more specimens in 74% of all cases). CNB was performed using an Alpha RT (Instrumentarium, Tuusula, Finland) film-based stereotaxy device. This guidance method required two technicians, US-guided CNB only one.

Costs and Diagnostic Value of CNB versus FNAB in Solid Breast Lesions

If a biopsy was considered necessary, it was taken immediately in an outpatient setting. If the first biopsy failed to yield a definite diagnosis, the need for a second biopsy using a different method was considered. Sometimes FNAB and CNB were performed simultaneously, e.g., if inflammatory carcinoma was suspected. After obtaining all the results, the radiologist decided whether the patient needed an operation, follow-up, or neither. The biopsy material was examined by pathologists at a private Finnish laboratory. Suspected or diagnosed cancer cases as well as some patients with benign tumors were sent to hospital for surgical treatment. If mammograms or US results were suspicious for malignancy, patients were sometimes referred to surgery even if the biopsies yielded negative results. Calculation of costs and statistical analysis The Statistical Package for Social Sciences (version 14.0; SPSS Inc., Chicago, Ill., USA) was used for all statistical analyses. Most variables were described using frequencies. All prices are given in Euro (t). The main concern here is with the costs incurred to the healthcare system; costs incurred to the patient were not considered. Calculations covered the costs of professionals (technicians, radiologists), cytological/ histological analysis of biopsies, radiological and other equipment (US, stereotaxy, Bard Magnum core biopsy needle [Bard Magnum, Covington, Ga., USA], Cameco syringe pistol [Precision Dynamics Corporation, Burbank, Calif., USA]), disposable supplies (needles, bandages, syringes, etc.), and social security and administrative costs (cleaning, building maintenance, etc.). Costs were calculated at 2005 prices, except for the technicians’ fees, which were only available for 2004. Costs for the services of radiologists and technicians were calculated on the basis of the average income and regular working hours for each occupational group in hospitals. The average amount of time used by radiologists and technicians for different biopsy methods was measured. Pathological analysis had fixed prices, which included postal charges. Costs from the use of the Cameco syringe pistol and Bard Magnum biopsy gun were calculated by dividing current purchase prices (including VAT) by their estimated service life (10 years). This 1-year price was then divided by the annual number of biopsies taken with that instrument. Costs from the use of ultrasound and stereotaxy were calculated in the same way, using the same estimated service life of 10 years. Current purchase price was based on

865

the average cost of a new US device suitable for breast examinations. The price of the stereotaxy device included the average cost of digital add-on stereotaxy plus 10% of a film-based mammography (including VAT). The decision to include only this 10% share was based on the fact that more mammographies than biopsies are usually performed with the equipment. Although our unit has film-based mammographystereotaxy apparatus, the decision was made to use the average cost of digital stereotaxy, which is becoming increasingly common and widespread. In the case of the US machine, the calculations allowed for the fact that it is used not only for biopsies but for other examinations as well. The total number of US examinations at our unit is currently around 2400 per year. To determine the costs of other equipment, we carefully calculated the de facto prices that had been paid for all syringes, bandages, and other small items. Social security costs were estimated by multiplying the technicians’ average income by a factor of 1.5. Administrative costs were estimated by multiplying the radiologist’s income by a factor of 1.7 (22). Establishing the accuracy of test results All cases with conservative treatment were followed up for 2 years after the biopsy or biopsies. After the follow-up, we searched the Finnish Cancer Registry to identify diagnosed cancer cases. If cancer was diagnosed in the biopsied breast during the followup, a negative biopsy result was considered a false negative. In the case of surgically treated patients, false-negative and -positive rates were established according to surgical outcome. Two patients referred to surgery were lost during the follow-up. FNAB result ‘‘suspicion of cancer’’ was always interpreted as either false negative or false positive (23). Results Almost one-quarter of the cases (n 156, 23%) showed no mammographic densities but only sonographically visible lesions. An enlarged axillary lymph node was found in 4% of cases in both mammography and US. Based on mammography findings, the cases were divided into modified BIRADS categories as follows: 112 (16%) normal, 140 (20%) benign, 152 (22%) low suspicion of malignancy, 34 (5%) high suspicion of malignancy, 220 (32%) malignant, and 30 (5%) not classified. Ultrasound was normal in 34 (5%) cases. Benign solid tumors were found in 188 (27%) cases, 92 Acta Radiol 2008 (8)

866

S.-M. Vimpeli et al.

(13%) findings were unspecific, and 374 (55%) were malignant or suspected of malignancy. Radiological analysis identified 575 (83%) solitary lesions. There were 66 (10%) multifocal lesions, 23 (3%) tumors were diffusely spread in the breast, and 12 (2%) of the cases had bilateral tumors. In 12 (2%) cases, information on foci was missing. The main results are shown in Tables 1
3. A total of 601 FNABs and 176 CNBs were performed during the study. In the FNAB group, 590 FNABs and 78 CNBs were performed; in the CNB group, 11 FNABs and 98 CNBs. The FNAB results were as follows: 12 (2%) insufficient sample, 47 (8%) normal, 232 (39%) benign, 53 (9%) suspicion of malignancy, 33 (5%) strong suspicion of malignancy, and 224 (37%) malignant. Overall, then, 51% of the FNAB results were malignant or suspicious. The CNAB results were as follows: five (3%) atypical ductal hyperplasia (ADH), 15 (8%) fibroadenoma, and 61 (35%) other benign finding. Other results included one (1%) lobular carcinoma in situ (LCIS), 15 (8%) ductal carcinoma in situ, 25 (14%) lobular invasive carcinoma, 47 (27%) ductal invasive carcinoma, four (2%) other malignancy, and three (2%) suspicion of malignancy. In total, 53% of the CNB findings were malignant (including three suspicious cases) and 47% benign (including ADH and LCIS). The false-negative rate for FNAB was 19%, and the false-positive rate 9%. The false-negative rate for CNB was 11%, and false-positive rate 1%. FNAB was most often guided by US, and CNB by stereotaxy. Neither biopsy method led to significant complications. Based on the results, 404 (58%) cases were recommended for surgery, but 12 of these patients (3%) did not have an operation either because of another disease, the nature of their breast cancer, or the patient’s refusal. The biopsies performed in both groups are listed in Table 4. FNAB was most expensive when performed under US guidance (t44.65), and CNB when Table 1. Number and percentage of mammographic density types. n

%

156 111 13 79 44 106 149 30 688

23 16 2 12 6 15 22 4 100

Table 2. Findings in mammography divided into modified BI-RADS categories.

Normal (1) Benign (2) Suspicion of malignancy (3) Strong suspicion of malignancy (4) Malignant (5) Information missing Total

n

% (n688)

112 140 152 34 220 30 688

16 20 22 5 32 5 100

performed under stereotaxic guidance (t246.37). FNAB was least expensive when performed with palpation (t36.51), and CNB likewise least expensive when performed with palpation (t70.02). Total biopsy costs in the FNAB group were t39,107, and in the CNB group t21,684. The biopsy costs per case were t66 in the FNAB group and t221 in the CNB group. The costs are listed in Table 5. Discussion Some of the patients in this study came from a screening program; others took part because of their clinical symptoms. In this sense, our sample was typical of normal practice. The common concern of these women was that they needed a biopsy assessment of a suspicious solid mass. Mammography showed that 37% of the cases were highly suspicious or definitely malignant. US showed that 55% were malignant or suspicious for malignancy. These rates obviously depended on the population chosen: they were lower than in studies conducted among cancer patients and higher than in studies with non-selected patients (2, 21). Most previous cost analyses have focused on comparing CNB with open biopsy. Their findings suggest that minimally invasive biopsies are more cost effective than the increasingly rare method of surgical biopsy (19, 24
29). The purpose of this study was to compare the costs of FNAB and CNB, a task that is complicated by the superior reliability of CNB in terms of both sensitivity and specificity (7, 11, 14
20). Cancer patients are often referred to Table 3. Number and percentage of different findings on US.

No densities Asymmetric density Architectural distortion Well-defined mass Lobulated mass Ill-defined mass Spiculated mass Information missing Total

Acta Radiol 2008 (8)

Normal Benign solid Nonspecific Suspicion of malignancy Malignant Total

n

%

34 188 92 143 231 688

5 27 13 21 34 100

867

Costs and Diagnostic Value of CNB versus FNAB in Solid Breast Lesions Table 4. Proportions of needle biopsies performed with different guidance methods in the FNAB and CNB groups. FNAB group FNAB performed FNAB guided by US FNAB guided by palpation FNAB guided by USpalpation CNB performed CNB CNB CNB CNB

guided guided guided guided

by by by by

CNB group

All

n 590

%

n 11

%

n 601

%

346 145 99

59 24 17

4 7 0

36 64 0

350 152 99

58 25 17

n 78

%

n 98

%

n 176

%

9 16 6 47

11 21 8 60

8 9 0 81

8 9 0 83

17 25 6 128

10 14 3 73

US palpation USpalpation stereotaxy

except two false-negative CNB cases were referred for surgery, as their imaging results were suspicious. This clearly underscores the importance of pooling all results and findings available, and also highlights the fact that all minimally invasive biopsy methods have their limitations (1, 7, 11, 14
20, 24). In FNAB, our insufficient sample rate was 2%, which is lower than in many earlier studies (4
46%) and probably reflects the experience of both radiologist and pathologist (12, 15). Reliable diagnosis by FNAB requires an experienced cytologist and good-quality samples. Many clinics today seem to have particular difficulties in finding an experienced cytologist (10, 11). Given the differences between the populations studied, it is very difficult to compare the results of FNABs. The cytological findings in our study were malignant (definite or suspicious) in 51% of all cases. This malignancy rate is lower than in some earlier reports (55
76%). One possible explanation is provided by our decision to include probably benign lesions (11, 24). Current guidelines by the US Agency of Health Care Policy and Research on the interpretive performance of mammography recommend that 25
40% of biopsies performed should prove to be malignant (30). At our imaging unit, often the first step with a tumor mass visible by US is to perform

surgery on the grounds of strongly suspicious or definitely malignant FNAB findings. CNB has the important advantage of allowing for histological analysis and the identification of hormone receptors, which is crucially important for preoperative treatment planning, especially with respect to sentinel node biopsy. Sometimes, of course, the radiologist does not have a choice between different biopsy methods. This is why cysts and abscesses were excluded from our study. For FNAB, the false-negative rate was 19% and false-positive rate 9%. For CNB, the false-negative rate was 11% and false-positive rate 1%. These figures clearly confirm that FNAB is a less reliable diagnostic method than CNB. The false-negative rate recorded for CNB here is higher than in most previous studies; the other figures are in line with expectations. The reasons for the high CNB falsenegative rate are not clear. What is clear, however, is that the staff at our unit will have to pay even closer attention to the quality of CNB biopsy techniques, equipment and pathological analysis, and possibly recover larger cores and larger numbers of cores. It has been suggested that the use of palpation as a guidance method may be the ultimate culprit, but in fact only one of the false-negative samples was obtained under palpation guidance. All

Table 5. Costs of needle biopsies, given in Euro and percentage of total price (VAT is included). US-guided FNAB

Radiologist Technicians Equipment Disposable supplies Pathological analysis Social security, administrative, etc., estimate Total

Palpationguided FNAB

Stereotaxyguided CNB

US-guided CNB

Palpationguided CNB

t

%

t

%

t

%

t

%

t

%

5.08 2.04 2.41 7.06 23.50

11 5 5 16 53

2.83 1.69 0.08 5.58 23.50

8 5 0 15 64

11.08 13.28 151.63 31.37 24.60

4 5 62 13 10

10.23 6.10 3.41 28.41 24.60

12

5.73

4 34 30

1.08 27.01 24.60

8 5.06 2 39 35

4.56 44.65

10 100

2.83 36.51

8 100

14.41 246.37

6 100

10.19 82.94

12 100

6.54 70.02

9 100

Acta Radiol 2008 (8)

868

S.-M. Vimpeli et al.

FNAB, which is a quick and easy method. If inflammatory carcinoma is suspected, both FNAB and CNB are sometimes performed simultaneously, as this is known to be a difficult diagnosis. With CNB, our malignancy rate was 53%. Earlier studies have reported widely varying figures, from 19 to 73% (11, 16, 19). There were no insufficient CNB samples. It has been recommended that CNB requires a 14G or larger needle. As mentioned above, the vacuum-assisted device offers some advantages, but it is also more expensive (10, 12, 14, 18). In the FNAB group, the cost of a diagnosis based on biopsies was t66; in the CNB group, the figure was more than three times this, at t221. The main reason why CNB is so expensive lies in the high costs of the stereotaxy device, as has been pointed out earlier (10, 12, 14, 16, 17, 19). In this study, 73% of CNBs were taken under the guidance of stereotaxy. Stereotactic guidance is often favored both in the literature and in clinical setups (19, 26
28, 30, 31), but it has been shown that CNB of solid lesions can be taken successfully under US guidance as well (12, 15
18, 32). According to our results, the costs of US-guided CNB were about one-third of the costs of stereotaxy-guided CNB. US guidance should be given preference whenever possible not only because of its lower costs, but also because it avoids ionizing radiation exposure. One way to keep the costs of stereotaxy in check is to provide a centralized service. On average, our imaging unit performs 14,060 mammographies, 2422 ultrasounds, 131 CNBs, and 403 FNABs a year. It is difficult to generalize our findings because income levels, administrative input, and many other factors vary widely from one country and even one hospital to the next (33). The biopsy costs in our study were very low compared to those reported earlier (19, 24, 28, 29). Most earlier calculations are from the United States, where both income levels and personnel numbers are higher than in Finland. Furthermore, the amount of time spent on one biopsy in Finland is shorter than indicated in some other countries (28). A useful way to compare the true costs of FNAB and CNB is to imagine a scenario where 1000 USguided FNABs are performed against 1000 USguided CNBs. Our results suggest that FNAB leads to an insufficient sample or an indeterminate result in 11% of cases, which need to be confirmed with CNB. In this scenario, 47% of FNAB results would be normal or benign which, according to some earlier reports, requires follow-up mammography (10, 11). Forty-two percent of FNAB results would Acta Radiol 2008 (8)

be malignant or strongly suspicious, and these cases would be referred for operation. According to our results, 47% of CNB results would be benign and referred for mammography control. Fifty-three percent of CNB results would be malignant and lead to surgery. In this scenario, the costs in the FNAB group are at least 20% lower than in the CNB group. For our assessment of cost effectiveness, we must also consider the reliability of FNAB and CNB. False-negative findings drive up the costs of treatment in more advanced stages of breast cancer. False-positive findings, for their part, lead to unnecessary operations that could have been avoided with more accurate diagnosis. All this means extra costs, especially in the FNAB group, and reduces the cost differential between the two groups. However, it was not possible to include these costs in our analysis, for that would have meant following up the treatment of each and every patient. As pointed out earlier, CNB offers the advantage of histological analysis and the identification of hormone receptors, which improves preoperative treatment planning and therefore probably lowers costs. In conclusion, both CNB and FNAB are less reliable than open surgical biopsy, and their results must be correlated with clinical and imaging findings. The follow-up of negative biopsy results should be considered. FNAB is a less reliable and less informative diagnostic method than CNB, and it is important to bear these limitations in mind. Even though negative or indeterminate FNAB results require follow-up or a re-biopsy with core needle, it is still a cost-effective procedure. Stereotaxy guidance considerably increases the costs of CNB. US guidance should be used whenever possible.

Acknowledgements We are very grateful to Eija Ja¨rvinen and Maisa Talviala for their invaluable help with data collection. The authors also wish to thank Professor Pekka Rissanen, Docent Prasun Dastidar, Ulla Ahokas, Marketta Heimola, Pa¨ivi Malkama¨ki, Eija Mannelin, Mari Niemi, Inga Nurmenniemi, Juhani Pakkanen, Pa¨ivi Rahko, Pa¨ivi Rantanen, Ari Talonen, Kristiina Va¨isa¨nen, and David Kivinen. Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

Costs and Diagnostic Value of CNB versus FNAB in Solid Breast Lesions

References 1. Berg WA, Gutierrez L, NessAiver MS, Carter WB, Bhargavan M, Lewis RS, et al. Diagnostic accuracy of mammography, clinical examination, US, and MR imaging in preoperative assessment of breast cancer. Radiology 2004;233:830
49. 2. Saarenmaa I, Salminen T, Geiger U, Heikkinen P, Hyvarinen S, Isola J, et al. Validity of radiological examinations of patients with breast cancer in different age groups in a population based study. Breast 2001;10:78
81. 3. Ahmed R, Ali SM. Role of imaging in diagnosis of carcinoma of breast. J Coll Physicians Surg Pak 2005; 15:238
41. 4. Rizzatto GJ. Towards a more sophisticated use of breast ultrasound. Eur Radiol 2001;11:2425
35. 5. Schepps B. Breast imaging: the role of alternative diagnostic modalities. Med Health R I 2005;88:80
1. 6. del Cura JL, Elizagaray E, Zabala R, Legorburu A, Grande D. The use of unenhanced Doppler sonography in the evaluation of solid breast lesions. Am J Roentgenol 2005;184:1788
94. 7. Parikh J, Tickman R. Image-guided tissue sampling: where radiology meets pathology. Breast J 2005;11: 403
9. 8. Graf O, Helbich TH, Fuchsjaeger MH, Hopf G, Morgun M, Graf C, et al. Follow-up of palpable circumscribed noncalcified solid breast masses at mammography and US: can biopsy be averted? Radiology 2004;233:850
6. 9. Hatmaker AR, Donahue RM, Tarpley JL, Pearson AS. Cost-effective use of breast biopsy techniques in a Veterans health care system. Am J Surg 2006;192: e37
41. 10. Russin LD. New directions in breast biopsy: review of current minimally invasive methods and presentation of a new coaxial technique. Semin Ultrasound CT MR 2000;21:395
403. 11. Meunier M, Clough K. Fine needle aspiration cytology versus percutaneous biopsy of nonpalpable breast lesions. Eur J Radiol 2002;42:10
6. 12. Helbich TH, Matzek W, Fuchsjager MH. Stereotactic and ultrasound-guided breast biopsy. Eur Radiol 2004;14:383
93. 13. Nasuti JF, Gupta PK, Baloch ZW. Diagnostic value and cost-effectiveness of on-site evaluation of fine-needle aspiration specimens: review of 5,688 cases. Diagn Cytopathol 2002;27:1
4. 14. Philpotts LE. Controversies in core-needle breast biopsy. Semin Roentgenol 2001;36:270
83. 15. Delle Chiaie L, Terinde R. Three-dimensional ultrasound-validated large-core needle biopsy: is it a reliable method for the histological assessment of breast lesions? Ultrasound Obstet Gynecol 2004;23:393
7. 16. Crystal P, Koretz M, Shcharynsky S, Makarov V, Strano S. Accuracy of sonographically guided 14-gauge core-needle biopsy: results of 715 consecutive breast biopsies with at least two-year follow-up of benign lesions. J Clin Ultrasound 2005;33:47
52. 17. Bolivar AV, Alonso-Bartolome P, Garcia EO, Ayensa FG. Ultrasound-guided core needle biopsy of nonpalpable breast lesions: a prospective analysis in 204 cases. Acta Radiol 2005;46:690
5.

869

18. Sauer G, Deissler H, Strunz K, Helms G, Remmel E, Koretz K, et al. Ultrasound-guided large-core needle biopsies of breast lesions: analysis of 962 cases to determine the number of samples for reliable tumour classification. Br J Cancer 2005;92:231
5. 19. Liberman L, Sama MP. Cost-effectiveness of stereotactic 11-gauge directional vacuum-assisted breast biopsy. Am J Roentgenol 2000;175:53
8. 20. Teh W. Intervention widens breast disease options. Diagnostic Imaging Europe 2006;February/March: 20
24, 63. 21. Zonderland HM, Pope TL Jr, Nieborg AJ. The positive predictive value of the breast imaging reporting and data system (BI-RADS) as a method of quality assessment in breast imaging in a hospital population. Eur Radiol 2004;14:1743
50. 22. Hujanen T. Guidelines for health care unit costs in Finland 2001. Helsinki, Finland: National Research and Development Centre for Welfare and Health; 2003; 1/2003, p. 86. 23. Riegelman RK, Hirsch P. Studying a study and testing a test. 5th ed. Philadelphia: Lippincott Williams & Wilkins; 2004. 24. Morris AM, Flowers CR, Morris KT, Schmidt WA, Pommier RF, Vetto JT. Comparing the cost-effectiveness of the triple test score to traditional methods for evaluating palpable breast masses. Med Care 2003; 41:962
71. 25. Hrung JM, Langlotz CP, Orel SG, Fox KR, Schnall MD, Schwartz JS. Cost-effectiveness of MR imaging and core-needle biopsy in the preoperative work-up of suspicious breast lesions. Radiology 1999;213:39
49. 26. Groenewoud JH, Pijnappel RM, van den Akker-Van Marle ME, Birnie E, Buijs-van der Woude T, Mali WP, et al. Cost-effectiveness of stereotactic large-core needle biopsy for nonpalpable breast lesions compared to openbreast biopsy. Br J Cancer 2004;90:383
92. 27. Fahy BN, Bold RJ, Schneider PD, Khatri V, Goodnight JE Jr. Cost-benefit analysis of biopsy methods for suspicious mammographic lesions. Arch Surg 2001;136: 990
4; discussion 994
5. 28. Bodai BI, Boyd B, Brown L, Wadley H, Zannis VJ, Holzman M. Total cost comparison of 2 biopsy methods for nonpalpable breast lesions. Am J Manag Care 2001;7:527
38. 29. Alonso-Bartolome P, Vega-Bolivar A, Torres-Tabanera M, Ortega E, Acebal-Blanco M, Garijo-Ayensa F, et al. Sonographically guided 11-G directional vacuum-assisted breast biopsy as an alternative to surgical excision: utility and cost study in probably benign lesions. Acta Radiol 2004;45:390
6. 30. Gur D, Wallace LP, Klym AH, Hardesty LA, Abrams GS, Shah R, et al. Trends in recall, biopsy, and positive biopsy rates for screening mammography in an academic practice. Radiology 2005;235:396
401. 31. Doyle AJ, King AR, Miller MV, Collins JP. Implementation of image-guided large-core needle biopsy of the breast on a limited budget. Australas Radiol 1998;42: 199
203. 32. Fenster A, Surry KJ, Mills GR, Downey DB. 3D ultrasound guided breast biopsy system. Ultrasonics 2004;42:769
74. 33. Klavus J, Linna M. International health care costcomparisons are problematic. Chess online 2006; 1/2006.

Acta Radiol 2008 (8)