Ultrasound-Guided Fine-Needle Aspiration of Clinically Negative Lymph Nodes Versus Sentinel Node Mapping in Patients at High Risk for Axillary Metastasis
Sonographically directed fine-needle aspiration is a less invasive and less costly alternative to sentinel node (SN) mapping in breast cancer patients at high risk for metastatic disease but with clinically negative axillae.
Radiographic, cytological, and histological diagnostic data on breast primary tumors from 114 consecutive SN candidates were prospectively assessed for clinicopathologic variables associated with an increased incidence of axillary metastases. Patients in whom these variables were identified underwent sonographic examination of their axillae followed by fine-needle aspiration when abnormal nodes were detected. SN mapping was performed in patients with normal axillary sonogram results or negative cytological results. Patients with positive cytological results proceeded to complete axillary dissection. Final axillary histological outcomes from patients not meeting the high-risk criteria were recorded. Additionally, a cost analysis was performed in which the costs of ultrasonography and ultrasound-guided fine-needle aspiration of the axilla were compared with those of SN mapping.
According to our selection criteria, a third of the patients with clinically negative axillae (37 of 114; 32%) were considered at high risk for axillary metastases. Fifty-nine percent of these patients (22 of 37) had metastatic disease on final histological analysis. Forty percent (15 of 37) of high-risk patients were spared SN mapping, with a reduction in health care costs of 20% in this patient population. Eighty-seven percent of patients not meeting high-risk criteria were SN negative.
This study suggests that in patients at increased risk for axillary metastases, the use of sonographic evaluation of the axilla in combination with fine-needle aspiration is not only clinically justified, but also cost-effective.
KeywordsUltrasound-guided cytology Breast cancer High-risk patients Sentinel node
A total of 30% to 40% of all breast cancer patients who undergo sentinel node mapping have nodal disease on final histological analysis.1, 2, 3, 4, 5, 6 Most of these patients have primary tumors with distinctive clinicopathologic features that are associated with an increased incidence of axillary metastasis.6, 7, 8, 9, 10 Although performing routine axillary ultrasound-guided fine-needle aspiration or core biopsy in all breast cancer patients is neither practical nor cost-effective,11, 12, 13, 14, 15, 16, 17, 18, 19 this subset of patients may benefit from additional evaluation of their axillae before sentinel node mapping. Our study focuses on the use of sonographic evaluation of the axillae as a screening test for high-risk patients and on the cost-effectiveness of ultrasound-guided fine-needle aspiration biopsy in this group of patients as a less invasive diagnostic alternative to sentinel node mapping.
Patient Selection Criteria
Patients with clinically negative axillae were considered at high risk for axillary nodal metastasis when the following characteristics were identified in their primary tumors at the time of initial diagnosis:6, 7, 8, 9, 10 grade III, size >1 cm (with or without lymphovascular invasion), or grade II, size >1.5 cm (with or without lymphovascular invasion). In histological specimens (needle core biopsies), grade was determined by using the modified Bloom-Richardson system.20 Cytological specimens were classified as high grade if the following features were present: nuclear pleomorphism, prominent nucleoli, and mitoses. The dimensions of the breast primary tumors were determined on the basis of their mammographic and/or sonographic size.
Criteria for Sonographically Abnormal Lymph Nodes
Individual Estimated Procedure Costs
The following estimates were provided by our hospital administration and include all hospital-related costs (not just the costs of the diagnostic procedures), with the exception of the physician charges. The charges do not represent actual insurance reimbursement. Ultrasonography cost $200, ultrasound-guided fine-needle aspiration biopsy cost $1250, sentinel node mapping cost $6300, and complete axillary node dissection cost $3700.
On final histological analysis, all positive sentinel lymph nodes in patients with either normal ultrasonography results (four patients) or negative fine-needle aspiration biopsy results (three patients) had at least 50% fatty replacement and a mean size of 1.2 cm (long axis). The average size of metastatic foci for these patients was 6.4 mm. This contrasts with the group of patients with abnormal ultrasound or positive fine-needle aspiration biopsy results, in whom the size of metastatic deposits was 11 mm for patients with a single positive node and 15 mm for patients with multiple positive nodes (Fig. 6).
According to present criteria, all the patients in this study were candidates for sentinel node mapping ($6,300 × 37 patients = $233,100). In retrospect, and on the basis of the final histological examination, 22 of them would have had complete axillary dissection ($3,700 × 22 patients = $81,400), for a total cost of $314,500 as a group or $8,500 per patient. According to our proposed algorithm, all patients underwent axillary ultrasonography ($200 × 37 patients = $7,400), and 22 patients had ultrasound-guided fine-needle aspiration biopsy ($1,250 × 22 = $27,500), 15 of whom were spared sentinel node mapping and proceeded directly to complete axillary dissection ($3,700 × 15 = $55,500). Twenty-two patients underwent sentinel node mapping ($6,300 × 22 = $138,600), seven with complete axillary node dissection ($3,700 × 7 = $25,900), for a total hospital cost as a group of $254,900 or $6,889.91 per patient (Fig. 4). This represents an approximately 20% overall hospital cost savings per patient.
Sentinel node mapping is currently the procedure with the greatest effect on diagnosis and treatment of breast cancer patients with clinically negative axillae.5,22,23 This is particularly true for patients with small (<pT1c) grade I or II hormone receptor–positive tumors, in which a negative sentinel node implies a minimal risk for future axillary recurrences and an effective measure for reducing the risk of lymphedema. Although it is a relatively simple procedure, sentinel node mapping is not without risk10,13,14 and may not be cost-effective in all breast cancer patients. When all patients undergoing sentinel node mapping are considered, the percentage of those with a high probability of axillary metastasis that will remain node negative on final histological analysis (15 of 114 [13%] in our series) is significantly lower than the percentage of patients with small nonaggressive lesions undergoing the same procedure (up to 60%).5,6 Our data suggest that for patients considered at high risk for axillary disease, introducing routine axillary ultrasonography, followed by fine-needle aspiration biopsy of abnormal nodes, could significantly reduce the number of sentinel node–mapping procedures (by 40% in this series), with an overall reduction in health care costs (20% according to our data) for this subset of patients.
Preoperative staging of breast cancer patients by using sonographic evaluation of the axilla with and without fine-needle aspiration biopsy is not a novel idea.11, 12, 13, 14, 15, 16, 17, 18, 19 Most studies are retrospective analyses of consecutive breast cancer series based solely on abnormal sonographic criteria, without targeting a specific patient population. These studies appropriately concluded that the indiscriminate use of sonographic examination of the axilla in all breast cancer patients, alone or in combination with fine-needle aspiration or core biopsies, was not practical or cost-effective.
Others have successfully explored the use of fine-needle aspiration biopsy alone or in combination with axillary ultrasonography and sentinel node mapping for preoperative chemotherapy patients and in locally advanced breast carcinoma for adequate axillary staging.24, 25, 26 These studies even recommend exploring the advantages of these techniques to stage the axilla in patients with smaller breast carcinomas. Our study addresses this last point precisely, studying the clinical and economic advantages of prospectively applying these diagnostic modalities to a targeted population on the basis of the pathologic characteristics of primary tumors that we associate with an increased likelihood of metastasis.
In most clinical practices, sonographic evaluation of new and/or mammographically suspicious palpable and nonpalpable breast masses is conducted as part of the routine diagnostic work-up for patients with potential breast carcinoma. Often this evaluation includes a sonographic survey of the corresponding axillary basin. In breast cancer centers with qualified cytopathologists (immediate on-site evaluation), the diagnostic assessment of the breast primary tumor and that of the abnormal node could occur as part of the same outpatient encounter. Consequently, a significant number of patients with clinically negative axillae who are at high risk for axillary disease could potentially have information regarding the need for standard axillary dissection versus sentinel node mapping in a single outpatient visit. In this study, a third of patients (7 of 21) with breast primary tumors diagnosed by fine-needle aspiration biopsy knew their positive axillary status and the need for complete axillary dissection at the end of a single outpatient clinic consultation. A similar scenario was observed during follow-up visits for patients with breast primary tumors diagnosed by needle core biopsies 1 to 2 weeks before their axillary sonographic examination (11 of 16 patients).
The lymph nodes with metastases missed by ultrasonography or fine-needle aspiration were different from those correctly diagnosed as positive by ultrasonography combined with fine-needle aspiration biopsy. Most patients with a false-negative/normal sonogram or false-negative fine-needle aspiration biopsy (86%; 6 of 7) had a single positive lymph node with tumor deposits having an average size of 8 mm and with at least 50% fatty replacement. This is in contrast to the group of patients with positive axillary fine-needle aspiration biopsies, in which 75% (10 of 15) had more than 3 lymph nodes involved by macrometastases, with a mean size of 16 mm. Although highly dependent on individual sonographic skills, these data suggest that metastatic disease involving a single node with small (<8 mm) metastatic deposits and a significant percentage of fatty replacement is more likely to be missed by these diagnostic modalities. However, according to our experience, patients with normal axillary ultrasound results or negative ultrasound-guided fine-needle aspiration biopsy of the axilla will remain node negative in 70% (15 of 22) of the cases or will have minimal axillary disease (single node involvement) in the reminder (Fig. 3).
These data become crucial when immediate breast reconstruction is considered, because most (70%) patients with cytologically proven node-positive breast cancers may be identified as having four or more metastatic axillary lymph nodes, which may be an indication for postmastectomy radiotherapy. Therefore, axillary ultrasonography is a reasonable strategy for “screening” the patients who may be better off with a delayed reconstruction as a result of the higher likelihood of postmastectomy chest wall radiation.
In contrast with previously published data,10,11 in our experience patients with breast primary tumors diagnosed by needle core biopsies did not generate postbiopsy reactive lymph node changes that resulted in a significant increase in abnormal sonographic axillary examinations and, thus, an increased number of unnecessary aspiration biopsies. The percentage of patients diagnosed as having abnormal nodes on sonographic examination but with negative fine-needle aspiration biopsy was similar for the group of patients with primary tumors diagnosed by fine-needle aspiration biopsy and for the group of patients with primary tumors diagnosed by needle core biopsy (9% [2 of 21] vs. 12% [2 of 16], respectively; Fig. 3).
Considering that this study was directed toward breast cancer patients with an anticipated higher incidence of axillary metastasis, the positive (82%) and negative (73%) predictive values, as well as the sensitivity (.82) and specificity (.73) figures for sonographic evaluation of the axilla and sonographically guided fine-needle aspiration biopsy (positive predictive value, 100%; negative predictive value, 57%; sensitivity, .83; specificity, 1), in our series are similar to, if not higher than, those previously reported by others with similar studies.10,12,14,15 All false-negative ultrasound cases shared specific clinicopathologic features: namely, they were single metastases <8 mm in small nodes <1 cm in diameter.
In this era of burgeoning health care costs, it is imperative that we maximize the accuracy, timeliness, and efficiency of diagnoses while avoiding unnecessary invasive, costly, and superfluous procedures. Although sentinel node mapping for the detection of axillary metastasis remains the gold standard, our data suggest that the use of sonographic evaluation of the axilla followed by fine-needle aspiration biopsy of abnormal nodes in patients with a higher risk for axillary metastasis fulfill these criteria and that this is not only clinically justified, but also cost-effective.
- 20.Elston C. Grading of invasive carcinoma of the breast. (1987) In: Page D, Anderson T, eds. Diagnostic Histopathology of the Breast. New York: Churchill Livingston, pp 300–11Google Scholar
- 21.Stavros T. (2003) Evaluation of regional lymph nodes in breast cancer patients. In: Breast Ultrasound. Philadelphia: Lippincott Williams & Wilkins, pp 855–64Google Scholar
- 26.Rouzier R, Extra JM, Klijanienko J, et al. Incidence and prognostic significance of complete axillary downstaging after primary chemotherapy in breast cancer patients with T1 to T3 tumors and cytologically proven axillary metastatic lymph nodes. J Clin Oncol 2002;20:1304–10PubMedCrossRefGoogle Scholar