Original Research

Doppler Ultrasound Evaluation of the Axilla in Clinically Node Negative Breast Cancer

Abhijit Das¹, Rahul Khanna¹, RN Meena¹, Ram C Shukla², Mohan Kumar³, Seema Khanna¹

• ¹Departments of Surgery, Institute of Medical Sciences, Banaras Hindu University, Varanasi – 221 005, India
• ²Departments of Radiology, Institute of Medical Sciences, Banaras Hindu University, Varanasi – 221 005, India
• ³Departments of Pathology, Institute of Medical Sciences, Banaras Hindu University, Varanasi – 221 005, India

• Submitted: May 20, 2012;
• Accepted June 19, 2012,
• Published: June 28, 2012

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Aim and Objective:

The study was conducted to evaluate the use of Doppler ultrasound in preoperative evaluation of clinically node negative axilla in patients with breast cancer and to correlate the findings with the tissue diagnosis obtained on subsequent sentinel lymph node biopsy or axillary dissection specimen.

Patients and Methods:

86 biopsy proven cases of breast cancer with clinically negative axilla were evaluated using 7.5 MHz high frequency ultrasound with Doppler. The L/S ratio, H/L ratio, flow pattern of lymph node, Resistivity Index (RI) and Pulsatility Index (PI) were recorded.

Results:

On the basis of the Doppler ultrasound findings it is possible to identify the subset of patients at high-risk of harboring nodal metastasis and these patients can be offered axillary nodal dissection upfront.

Conclusions:

Patients who have an equivocal or indeterminate or negative axilla on Doppler ultrasound should be advised to undergo sentinel lymph node biopsy prior to decisions regarding further management of the axilla.

Introduction

Accurate assessment of the axillary lymph node status is crucial in the treatment of breast cancer. Sentinel lymph node biopsy (SLNB) evaluation has become the standard of care for the management of axilla in clinically node negative breast cancer patients. Further therapeutic decisions regarding the need for axillary lymph node dissection are made intra-operatively on the basis of the status of the sentinel node.

The morbidity associated with SLNB although less than that of axillary lymph node dissection is still considerable. This includes the increased cost and complexity of the surgical procedure. SLNB requires the availability of a radioactive colloid, facilities of lymphoscintiscan and that for frozen section analysis. It is also associated with a false negative rate of 5-10% [1]. SLNB is not reliable in patients with axillary disease or with a high risk of disease positive nodes because of the increased chances of lymphatic blockade leading to harvesting of healthy ‘false sentinel nodes’ [2].

The aim of our study was to use Doppler ultrasound for the preoperative evaluation of clinically node negative axilla in breast cancer and to correlate the findings with the tissue diagnosis obtained on subsequent SLNB or axillary dissection specimen. The goal was to identify that subset of patients who could be reliably diagnosed on the basis of Doppler USG to harbor axillary nodal metastasis. Such patients could be spared a SLNB procedure and offered ALND upfront.

Table 1: Doppler ultrasound findings in 86 patients of clinically node negative breast cancer

Color Doppler parameters	Mean+SD	Range
No. of lymph nodes	2.04 + 0.88	1 to 6
L/S Ratio	1.70 + 0.55	1.01 to 2.82
H/L Ratio	0.46 + 0.28	0 to 1.21
Resistivity Index	0.81 + 0.15	0.56 to 1.06
Pulsatility Index	1.88 + 0.80	0.77 to 4.29

Patients and Method

This prospective clinical study was carried out over a 4 years period (July 2007 to June 2011) in the Department of Surgery, Institute of Medical Sciences, Banaras Hindu University, Varanasi. Included in the study were 86 FNAC proven breast cancer patients with a clinically negative axilla. Patients with history of prior chemotherapy or radiotherapy, breast biopsy, prior axillary intervention, pregnant and lactating women, and patients with non visualization of SLN by blue dye technique during surgery were excluded.

The study was approved by the Institutional Ethics Committee and written informed consent was obtained from all patients. All patients underwent a thorough clinical examination including careful palpation of both axillae. Tissue diagnosis of breast cancer was obtained by FNAC and in equivocal cases was confirmed by trucut^® needle biopsy.

Doppler ultrasound evaluation of the ipsilateral axilla was done using high resolution real time Toshiba nemio premium compact ultrasound device with a high frequency linear array probe of 7.5 MHz. The patients were examined in supine position with the arm in 900 abduction and external rotation. In this position the axillary vessels have a nearly straight course facilitating orientation and all parts of the axilla can be thoroughly examined. The pectoral and infraclavicular region around the subclavian vessels were also scanned. The following parameters were recorded:

•  L/S ratio : long axis / short axis of lymph node
•  H/L ratio : long axis of hilum / long axis of lymph node
•  Flow pattern of lymph node : central or peripheral
•  Resistivity Index (RI) : Peak systolic velocity – End diastolic velocity / Peak systolic velocity
•  Pulsatility Index (PI) : Peak systolic velocity – End diastolic velocity / Mean velocity

All the nodes were assessed on the above parameters. In axillae with multiple nodes, the node with minimum L/S ratio, H/L ratio and maximum RI and PI were taken into account.

All patients underwent modified radical mastectomy sparing the Pectoralis minor tendon. During the surgery peroperative sentinel lymph node localization was done using the blue dye technique. The sentinel node was harvested and subjected to Touch Imprint Cytology (TIC). The mastectomy and axillary specimen along with the harvested SLN were evaluated by conventional histopathology. The results of Doppler ultrasound were correlated with the status of the SLN and also with the histopathology of the non sentinel lymph nodes harvested in the axilla.

Table 2: Correlation of histopathological status of SLN and Doppler ultrasound findings in 86 patients

SLN       status on HPE	No. of LNs identified	L/S Ratio	H/L Ratio	RI	PI	Blood Fluid pattern of LN
Positive (n=32)	2.81 + 0.60	1.21 + 0.15	0.24 + 0.23	0.96 + 0.07	2.46 + 0.88	Peripheral : 26 Central : 6
Negative (n=54)	1.42 + 0.51	2.0 + 0.43	0.64 + 0.17	0.69 + 0.07	1.42 + 0.29	Peripheral : 7 Central : 47
p value	0.67	0.004	0.025	0.906	0.004	0.01

Results

All the 86 patients included in the study were female and their mean age was 52 years (range 34 – 71 years). On the basis of tumor size the patients were classified as T1: 31 (36%), T2: 50 (58%) and T3: 5 (6%). As per protocol all patients had a clinically negative axilla. The Doppler ultrasound findings of the axilla in these patients are depicted in Table 1.

At the time of modified radical mastectomy, it was possible to identify the SLN using the blue dye technique in 86 patients out of 94 in whom it was attempted (sensitivity of 92%). On TIC of the SLN, there was evidence of metastasis in 28 patients (32%) while 58 patients (68%) were free of tumor deposits. Subsequent histopathology of these 86 sentinel nodes found evidence of metastasis in 32 patients (37%) while 54 patients (63%) had negative sentinel nodes. Thus the sensitivity and specificity of TIC viz a viz histopathology of the SLN was found to be 88% and 100% respectively. On histopathology of the non SLN, evidence of metastasis was found in 14 patients while in 18 patients the SLN was the only site of metastasis. Significantly in none of the 58 patients who had a negative SLN on TIC was a deposit found in non SLN.

Correlation was made between the histopathological status of the SLN and the Doppler ultrasound findings which are depicted in Table 2.

Discussion

Sentinel node biopsy has now become the standard therapeutic modality for clinically node negative axilla in breast cancer patients. The aim is to avoid axillary dissection in patients in whom the sentinel node is found to be free of metastasis. However sentinel node biopsy has not found wide acceptance in developing countries like India due to several reasons. These are requirement of nuclear medicine and frozen section facilities, waiting intraoperatively for frozen section reports thus prolonging operating time and a 5-10% false negative rate.

Table 3: Accuracy of various Doppler – ultrasound parameters for evaluation of metastatic axillary lymph nodes in comparison to histopathology of the sentinel node

Doppler – USG Parameter	Sensitivity	Specificity	Positive Predictive Value	Negative Predictive Value	Diagnostic Accuracy
Number of LN 3 or more	69%	85%	73%	82%	79%
L/S ratio <1.4	87%	89%	82%	92%	88%
H/L ratio <0.30	81%	83%	74%	88%	82%
RI >0.90	59%	64%	50%	73%	63%
PI >2.2	81%	78%	68%	87%	79%
Peripheral flow pattern	62%	67%	52%	75%	65%
All the above parameters	59%	65%	50%	73%	63%

Differentiation between metastatic and non metastatic axillary lymph nodes is extremely important from the therapeutic view point. It is also important to make the correct diagnosis at the earliest because a delayed diagnosis can lead to upstaging of the disease making a curable lesion incurable. Given the limitations of SLNB, we decided to evaluate the efficacy of Doppler ultrasonogrpahy in differentiating metastatic from non metastic nodes in clinically node negative breast cancer.

Doppler ultrasonography can evaluate various parameters like ratio of long axis to short axis (L/S ratio), hilum axis to long axis ratio (H/L ratio), vascular pattern, displacement of vascularity, vascular resistance (RI) and pulsatility index (PI). These features have been reported to have a sensitivity of 88% for the diagnosis of metastatic nodes and 67% for lymphoma with a specificity of 100% [3]. Normal lymph nodes are generally depicted as low echogenic oval structures with a clear visible hilum containing the central vessels. Lymph node metastasis are characterized by the following sonomorphological changes: increase in size (short diameter > 8 mm, ratio of length / width <2) and number, change in shape (more round, polygonal) and sonotexture (cystic, necrotic inclusions, hypoechogenic appearance, disappearance of hilar region) [4]. These changes can be easily detected by high frequency linear transducers.

Long axis / Short axis Ratio (L / S Ratio)

The size of the lymph node is considered to be an important parameter to distinguish benign from malignant nodes. But lymph node enlargement could be malignant on or inflammatory. Therefore it is not just the long axis of lymph nodes but their globular shape which helps to distinguish benign from malignant nodes. The globular shape can be assessed by the L/S ratio. Khanna R et al., in their study of cervical lymph nodes described the L/S ratio to be 2.2 in benign nodes [5]. Sakai et al., described ‘Rounding Index’ to be >2.0 to describe a node as benign [6]. In the present study the L/S ratio if metastatic lymph nodes was 1.21+0.15 while that of non-metastatic nodes was 2.0+0.43 (p value = 0.004). L/S ratio of <1.4 had a sensitivity of 87%, specificity of 89% and a diagnostic accuracy of 88% for the diagnosis of metastatic lymph node.

Hilum axis / Long axis Ratio (H/L Ratio)

The hilum of a normal lymph node constitutes the central part of the lymph node and appears homogeneously hyperechoic on Doppler ultrasound. Metastatic deposit leads to shrinkage of hilar area and loss of echogenicity on sonography. Hilar diameter to long axis ratio has been used to distinguish benign from malignant nodes. Nori et al., found H/L ratio <0.50 to be highly suspicious of malignant deposit [7]. We found that the H/L ratio of malignant nodes was 0.245 +0.23 and those of benign nodes was 0.644 + 0.17 (p value = 0.025). An H/L ratio <0.30 was found to predict malignant deposit in the node with a sensitivity of 81%, specificity of 83% and a diagnostic accuracy of 82%.

Resistivity Index (RI)

The principle of RI is that in a lymph node, the high resistance of distal vessels produces a low diastolic flow in the feeding artery increasing the difference between peak systolic and end diastolic velocity. Ho et al., have described RI >0.8 to label a node malignant [3]. Ahuja et al., observed that 54% of malignant nodes and 4% of reactive nodes had an RI >0.8 which would result in a low sensitivity but relatively high specificity [8]. We observed that RI of metastatic nodes was 0.96 + 0.07 and that of non metastatic nodes was 0.69 + 0.07 (p value = 0.906). The difference was non significant and was not found to be useful for differentiating benign from metastatic nodes.

Pulsatility Index (PI)

This is a measure of the variability of blood flow in a vessel and is equal to the difference between peak systolic and minimal diastolic velocity divided by the mean velocity during the cardiac cycle. Ahuja et al., found 74% of metastatic and 6% of reactive nodes had a PI > 1.5 yielding a high specificity but low sensitivity [8]. Ho et al., described PI > 1.5 to differentiate metastatic from non metastatic nodes [3]. In the present study the PI of metastatic nodes was 2.46 + 0.88 and that of non metastatic nodes was 1.42 + 0.29 (p value = 0.004). A PI > 2.2 was found to diagnose metastatic nodes with a sensitivity of 81%, specificity of 78% and a diagnostic accuracy of 79%.

Arterial Flow Pattern

The normal blood flow to lymph nodes is through the hilum which gets intensified during inflammation. In case of metastatic deposits the hilum is disrupted and the flow gradually shifts to the peripheri. We found that out of 32 metastatic axillary nodes, peripheral flow pattern was seen in 26 (81%) while in the 54 normal axillae, peripheral flow pattern was observed in only 7 (13%) cases (p=0.01). Therefore a peripheral flow pattern was found to diagnose metastatic nodes with a sensitivity of 62%, specificity of 67% and a diagnostic accuracy of 65% overall.

Ultrasound evaluation has certain inherent limitations. It is operator dependant and the interpretation may vary from examiner to examiner. Visualization of lymph nodes may be hampered by overlying fat in obese subjects. There may be technical problems like inaccurate gray – scaling and limited tissue penetration. Most importantly differentiation between metastatic and reactive inflammatory nodes may not be possible by ultrasonography. Addition of duplex sonography may be helpful in such situations.

Flow patterns such as central or peripheral, velocity ratios like pulsatility index (PI, reflecting peripheral vessel obstruction) and resistivity index (RI, reflecting increased peripheral arterial resistance) can help differentiate metastatic from reactive lymph nodes [9]. Metastic nodes have a predominantly peripheral blood flow pattern due to displacement of the physiological nodal vessels by the metastatic deposit while in reactive lymph nodes, the flow is mostly central [10]. Similarly RI and PI values are higher in metastatic nodes as compared to reactive nodes.

The most important draw back of axillary ultrasound is its imability to detect micrometastasis [11]. These micrometastasis could be detected on a sentinel node biopsy especially if a immuno-histochemical examination is done in addition to the frozen section analysis. The implication is that patients who have a suspicious axillary ultrasound can be offered an axillary nodal dissection upfront without the need of a sentinel node biopsy. However, patients with a normal axillary ultrasound need a sentinel node biopsy to confirm their negative axilla status.

Conclusions

This prospective study was carried out on 86 breast cancer patients with a clinically negative axilla who underwent a Doppler ultrasound evaluation of their axilla prior to surgery. It was found that LN number >3, L/S ratio <1.4, H/L ratio <0.30 and PI >2.2 were most reliable parameters for diagnosing metastatic lymph nodes with a diagnostic accuracy of 79%, 88%, 82% and 79% respectively. The limitation of axillary ultrasound is its inability to detect micrometastasis which is better diagnosed on sentinel node biopsy. On the basis of Doppler ultrasound findings it is possible to identify the subset of patients at high risk of harboring nodal metastasis and these patients can be offered axillary nodal dissection upfront. Patients who have an equivocal or indeterminate or negative axilla on Doppler ultrasound should be advised to undergo sentinel lymph node biopsy prior to decisions regarding further management of the axilla.

Conflicts of Interests

The authors declare that there are no conflict of interests

Authors’ contribution

AD: Data collection, writing of manuscript
RK: Design of study, editing the paper
RNM: Literature search and data collection
RCS: edited the radiological part of the manuscritp
MK: edited the patholgical part of the manuscript
SK: Data collection, analysis, editing

Ethical Considerations

The study was approved by the Institute Ethics Committee.

Funding

None

Acknowledgements

None

References

[1]. Gill G; SNAC Trial Group of the Royal Australasian College of Surgeons (RACS) and NHMRC Clinical Trials Centre.. Sentinel-lymph-node-based management or routine axillary clearance? One-year outcomes of sentinel node biopsy versus axillary clearance (SNAC): a randomized controlled surgical trial. Ann Surg Oncol. 2009; 16: 266 – 275.

[2]. García Fernández A, Fraile M, Giménez N, Reñe A, Torras M, Canales L, Torres J, Barco I, González S, Veloso E, González C, Cirera L, Pessarrodona A. Use of axillary ultrasound, ultrasound-fine needle aspiration biopsy and magnetic resonance imaging in the preoperative triage of breast cancer patients considered for sentinel node biopsy. Ultrasound Med Biol. 2011 Jan;37(1):16-22.

[3]. Ho SS, Metreweli C, Ahuja AT. Does anybody know how we should measure Doppler parameters in lymph nodes? Clin Radiol. 2001; 56: 124 – 26.

[4]. Wunder baldinger P. Problems and prospects of modern lymph node imaging. Eur J Radiol 92006): 58; 325 – 337.

[5]. Khanna R, Sharma AD, Khanna S, Kumar M, Shukla RC. Usefulness of ultrasonography for the evaluation of cervical lymphadenopathy. World Journal of Surgical Oncology (2011) : 9; 29.

[6]. Sakai F, Kigono K, Sone S, et al. Ultrasonic evaluation of cervical metastatic lymphadenopathy. J Ultrasound Med (1998) : 7; 305 – 10

[7]. Nori J. Vanizi E, Bazzocchi M, Bufalini F, Distante V, Branconi F, Susini T. Role of axillary ultrasound examination in the selection of breast cancer patients for sentinel node biopsy. American Journal of Surgery (2007); 193; 16 – 20.

[8]. Ahuja AT, Ying M, Ho SS, Metreweli C. Distribution of intranodal vessels in differentiating benign from metastatic neck nodes. Clinical Radiology (2001) : 56 : 197 – 201.

[9]. Tschammler A, Wirkner H, Ott G, Hahn D. Vascular pattern in reactive and malignant lymphadenopathy. Eur Radiol (1996): 6: 473 – 80.

[10]. Solbiate L, Osti V, Cova L, Tonolini M. Ultrasound of thyroid, parathyroid glands and neck lymph nodes. Eur Radiol (2001): 11 (12): 2411 – 24.

[11]. Susini T, Nori J, Olivieri S et al. Predicting the status of axillary lymph nodes in breast cancer: A multiparameter approach including axillary ultrasound scanning. The Breast (2009): 16; 103-108.