DOI:10.2214/AJR.07.2460
AJR 2008; 191:646-652
© American Roentgen Ray Society
Cortical Morphologic Features of Axillary Lymph Nodes as a Predictor of Metastasis in Breast Cancer: In Vitro Sonographic Study
Deepak G. Bedi1,
Rajesh Krishnamurthy2,
Savitri Krishnamurthy3,
Beth S. Edeiken1,
Huong Le-Petross1,
Bruno D. Fornage1,
Roland L. Bassett, Jr.4 and
Kelly K. Hunt5
1 Division of Diagnostic Imaging, The University of Texas M. D. Anderson Cancer
Center, 1515 Holcombe Blvd., Unit 1350, Houston, TX 77030.
2 Department of Diagnostic Imaging, Texas Children's Hospital, Houston,
TX.
3 Department of Pathology, The University of Texas M. D. Anderson Cancer Center,
Houston, TX.
4 Division of Quantitative Sciences, The University of Texas M. D. Anderson
Cancer Center, Houston, TX.
5 Department of Surgical Oncology, The University of Texas M. D. Anderson Cancer
Center, Houston, TX.
Received April 24, 2007;
accepted after revision March 9, 2008.
Address correspondence to D. G. Bedi.
Abstract
OBJECTIVE. The purpose of this study was in vitro
sonographic–pathologic correlation of findings in dissected axillary
lymph nodes from breast cancer patients undergoing axillary lymph node
dissection and classification of the sonographic appearance of the nodes on
the basis of cortical morphologic features to facilitate early recognition of
metastatic disease.
MATERIALS AND METHODS. High-resolution sonography was used for in
vitro examination of 171 lymph nodes from 19 axillae in 18 patients with
unknown nodal status who underwent axillary lymph node dissection for early
infiltrating breast cancer. The images were evaluated by two blinded
observers, and discordant readings were referred to a third blinded observer.
Each lymph node was classified as one of types 1–6 according to cortical
morphologic features. Types 1–4 were considered benign, ranging from
hyperechoic with no visible cortex to thickened generalized hypoechoic
cortical lobulation. Type 5 (focal hypoechoic cortical lobulation) and type 6
(hypoechoic node with absent hilum) nodes were considered metastatic. The
reference standard for metastatic disease was histopathologic evaluation of
sectioned nodes by a single pathologist blinded to sonographic findings.
Largest nodal diameter also was measured.
RESULTS. Interobserver agreement was 77% for classification of nodal
morphology (types 1–6) and 88% for characterization of a node as benign
or malignant. Sensitivity, specificity, positive predictive value, negative
predictive value, and overall accuracy of cortical shape in prediction of
metastatic involvement of axillary nodes were 77%, 80%, 36%, 96%, and 80%.
Type 4 nodes had the most false-negative findings (four of 36). Node size
ranged from 0.2 to 3.8 cm, and subcentimeter nodes of all types were
detected.
CONCLUSION. In breast cancer, axillary lymph nodes can be classified
according to cortical morphologic features. Predominantly hyperechoic nodes
(types 1–3) can be considered benign. Generalized cortical lobulation
(type 4) is uncommonly a false-negative finding, but metastasis, if present,
is invariably detected at sentinel node mapping. The presence of asymmetric
focal hypoechoic cortical lobulation (type 5) or a completely hypoechoic node
(type 6) should serve as a guideline for universal performance of fine-needle
aspiration for preoperative staging of breast cancer. This classification,
when verified with larger samples, may serve as a useful clinical guideline if
proven with results of in vivo studies.
Keywords: breast cancer • nodal metastasis • nodal sonography
Introduction
The status of axillary and other regional lymph nodes is vital in the
initial staging of breast cancer
[1,
2]. Physical examination and
mammography are not sufficiently accurate
[3,
4] for detecting nodal
metastasis. A growing body of literature now recognizes sonography combined
with fine-needle aspiration (FNA) as the most useful means of preoperative
evaluation of the axilla for the presence of nodal metastasis
[5–11].
Although the overall accuracy of sonography has been high, variable technique
and differing criteria for nodal malignancy have occasionally resulted in
mixed performance in studies
[5]. The sonographic criteria
for metastasis have been focused on node shape (length-to-width ratio) and
overall echogenicity (presence or absence of hyperechoic hilum). Focal changes
in the cortical morphologic features of a node may be more important because
metastatic cells are first deposited in the periphery of a node
[12] (Figs.
1,
2,
3). Although eccentric cortical
enlargement has been mentioned as a criterion for malignancy
[7], this feature has not been
sufficiently explored. A few in vitro sonographic studies of dissected nodes
have been focused mainly on nodal shape
[13–16]
and may have been limited by early equipment. Results with in vitro sonography
of dissected axillary nodes in a study by Feu et al.
[13] and with helical CT
obtained in a study by Uematsu et al.
[17] support cortical
morphology as an important feature.
View larger version (48K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1 —Diagram shows anatomic features of lymph node. Hypoechoic
cortex (C) on sonogram represents marginal sinus, lymphoid follicles (F), and
paracortex (P). Paracortex occasionally is slightly more hyperechoic
(Fig. 3) owing to fat
infiltration from hilum (Fig.
7B). Hilum (H) is hyperechoic owing to multiple reflective
interfaces of blood vessels, fat, and central sinus.
|
|
View larger version (97K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3 —33-year-old woman with infiltrating ductal cancer. Sonogram
of benign lymph node shows hypoechoic cortex (C) with slightly hyperechoic
paracortex (P) can be correlated to outer zone of lymphoid tissue in
Figure 1. Hilum (H) is
hyperechoic, representing central sinus, medullary cords, blood vessels, and
fat. Afferent and efferent vessels cannot be seen with sonography.
|
|
Our objective was to perform in vitro sonographic–pathologic
correlation of findings on dissected axillary lymph nodes from breast cancer
patients undergoing axillary lymph node dissection and to classify the
sonographic appearance of the nodes on the basis of cortical morphologic
features to facilitate recognition of metastatic deposits before complete
replacement of a node.
Materials and Methods
The study was performed at a major cancer institution by physicians
performing only oncologic work who had at least 8 years to more than 20 years
of experience. Patients were consecutively recruited into the study over a
3-week period (January–February 2001) if they were determined to have
invasive breast cancer necessitating axillary lymph node dissection and if the
nodal status was unknown. Patients with nodal metastasis from preoperative
sonographically guided FNA were excluded. Sentinel lymph nodes bio p sied
during a surgical procedure were not scanned and were excluded to avoid delays
in diagnosis while a patient was under anesthesia. Patients with a history of
neoadjuvant chemo therapy were included. Institutional review board approval
was obtained for a laboratory protocol because all studies were performed on
resected tissue rather than live patients. The 18 patients described were part
of an initial study of the feasibility of repeating the protocol with a larger
sample.
In vitro sonography was performed on 186 dissected lymph nodes from 19
axillae in 18 patients undergoing axillary lymph node dissec tion as part of
breast cancer staging and treat ment. The dissected axillary specimen was de
livered to the pathology laboratory in the operat ing room, where a
pathologist carefully cut it into smaller pieces on the basis of palpation of
nodes. The pieces were placed into saline solution in a labeled com
partmentalized neutralized plastic tray, each section being large enough to
allow scanning of the cut pieces with a sonographic transducer. Real-time
sonography of each section of tissue was performed in the pathology labora
tory with a sonographic unit with 10- to 12-MHz linear-array trans ducers. A
surgical forceps was sometimes used to immobilize smaller pieces of tissue to
allow accurate cross-sectional imaging. When more than one node was seen
within a piece of tissue, the piece was further subdivided care fully with a
scalpel so that only one node was seen per piece. When no nodes were seen in a
piece of tissue, this finding was recorded as no node and sent for histologic
evaluation. Imaging was recorded on videotape to enable review of real-time
images for selection of the true cross-section of a node. The total elapsed
scanning time for each patient was 20–40 minutes, depend ing on the
number of sections available. The tissue was returned to the pathologist for
histologic examination.
View larger version (93K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 7B —52-year-old woman with invasive intraductal carcinoma and
benign type 4 node. Sonogram (A), histopathologic photograph
(B), and diagram (C) show cortical lobulations (arrows)
are generalized and follow contour of hilar echogenicity. Cortex appears
thinner on A because of hilar fat infiltration into paracortex
(arrowheads, A and B). H = hilum.
|
|
View larger version (112K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4A —54-year-old woman with invasive ductal carcinoma and benign
node (type 1). Sonogram (A), histopathologic photograph (B), and
diagram (C) show almost no cortex (arrows), which is more
evident in B. Hilum (H) has paradoxically hypoechoic areas due to
presence of relatively few vessels and mostly homogeneous fat cells and lacks
many reflective interfaces. Relatively hyperechoic areas correlate with
vessels and trabeculae.
|
|
View larger version (128K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4B —54-year-old woman with invasive ductal carcinoma and benign
node (type 1). Sonogram (A), histopathologic photograph (B), and
diagram (C) show almost no cortex (arrows), which is more
evident in B. Hilum (H) has paradoxically hypoechoic areas due to
presence of relatively few vessels and mostly homogeneous fat cells and lacks
many reflective interfaces. Relatively hyperechoic areas correlate with
vessels and trabeculae.
|
|
View larger version (6K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4C —54-year-old woman with invasive ductal carcinoma and benign
node (type 1). Sonogram (A), histopathologic photograph (B), and
diagram (C) show almost no cortex (arrows), which is more
evident in B. Hilum (H) has paradoxically hypoechoic areas due to
presence of relatively few vessels and mostly homogeneous fat cells and lacks
many reflective interfaces. Relatively hyperechoic areas correlate with
vessels and trabeculae.
|
|
View larger version (97K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5A —54-year-old woman with invasive ductal carcinoma and benign
type 2 node. Sonogram (A), histopathologic photograph (B), and
diagram (C) show uniform thin hypoechoic cortex (arrows) less
than 3 mm thick. B shows thin cortex around hilum (H).
|
|
View larger version (60K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5B —54-year-old woman with invasive ductal carcinoma and benign
type 2 node. Sonogram (A), histopathologic photograph (B), and
diagram (C) show uniform thin hypoechoic cortex (arrows) less
than 3 mm thick. B shows thin cortex around hilum (H).
|
|
View larger version (8K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5C —54-year-old woman with invasive ductal carcinoma and benign
type 2 node. Sonogram (A), histopathologic photograph (B), and
diagram (C) show uniform thin hypoechoic cortex (arrows) less
than 3 mm thick. B shows thin cortex around hilum (H).
|
|
View larger version (108K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 6A —49-year-old woman with invasive ductal carcinoma and benign
type 3 node. Sonogram (A), histopathologic photograph (B), and
diagram (C) show uniform cortex thicker than 3 mm and minor surface
lobulations (arrows, A and B). Hyperechoic hilum (H)
has more reflective interfaces than node in Figure
4A,
4B,
4C.
|
|
View larger version (142K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 6B —49-year-old woman with invasive ductal carcinoma and benign
type 3 node. Sonogram (A), histopathologic photograph (B), and
diagram (C) show uniform cortex thicker than 3 mm and minor surface
lobulations (arrows, A and B). Hyperechoic hilum (H)
has more reflective interfaces than node in Figure
4A,
4B,
4C.
|
|
View larger version (10K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 6C —49-year-old woman with invasive ductal carcinoma and benign
type 3 node. Sonogram (A), histopathologic photograph (B), and
diagram (C) show uniform cortex thicker than 3 mm and minor surface
lobulations (arrows, A and B). Hyperechoic hilum (H)
has more reflective interfaces than node in Figure
4A,
4B,
4C.
|
|
The real-time imaging videotape was reviewed separately by two radiologists
blinded to each other's findings. Each node was classified into one of the
following types on the basis of cortical morphologic features: type 1,
hyperechoic, no visible cortex (Fig.
4A,
4B,
4C); type 2, thin (< 3 mm)
hypoechoic cortex (Fig. 5A,
5B,
5C); type 3, hypoechoic cortex
thicker than 3 mm (Fig. 6A,
6B,
6C); type 4, generalized
lobulated hypoechoic cortex (Fig.
7A,
7B,
7C); type 5, focal hypoechoic
cortical lobulation (Figs. 8A,
8B,
8C); type 6, totally
hypoechoic node with no hilum (Figs.
8D,
8E,
8F). Nodes were empirically
considered benign (types 1–4) or suspicious or metastatic (types 5 and
6) on the basis of experience
[6] and anatomic knowledge of
nodes. Discordant findings were referred to a third blinded observer. The
reference standard for metastatic disease was histopathologic features of the
nodes determined by a pathologist blinded to the sonographic findings.
View larger version (108K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 7A —52-year-old woman with invasive intraductal carcinoma and
benign type 4 node. Sonogram (A), histopathologic photograph
(B), and diagram (C) show cortical lobulations (arrows)
are generalized and follow contour of hilar echogenicity. Cortex appears
thinner on A because of hilar fat infiltration into paracortex
(arrowheads, A and B). H = hilum.
|
|
View larger version (10K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 7C —52-year-old woman with invasive intraductal carcinoma and
benign type 4 node. Sonogram (A), histopathologic photograph
(B), and diagram (C) show cortical lobulations (arrows)
are generalized and follow contour of hilar echogenicity. Cortex appears
thinner on A because of hilar fat infiltration into paracortex
(arrowheads, A and B). H = hilum.
|
|
View larger version (95K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 8A —42-year-old woman with invasive ductal carcinoma and
metastatic type 5 and 6 nodes. Sonogram (A), low-power histopathologic
photograph (B), and diagram (C) show type 5 node characterized
by focal hypoechoic lobulation of cortex (arrows) due to metastatic
deposit (T). H = hilum.
|
|
View larger version (68K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 8B —42-year-old woman with invasive ductal carcinoma and
metastatic type 5 and 6 nodes. Sonogram (A), low-power histopathologic
photograph (B), and diagram (C) show type 5 node characterized
by focal hypoechoic lobulation of cortex (arrows) due to metastatic
deposit (T). H = hilum.
|
|
View larger version (9K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 8C —42-year-old woman with invasive ductal carcinoma and
metastatic type 5 and 6 nodes. Sonogram (A), low-power histopathologic
photograph (B), and diagram (C) show type 5 node characterized
by focal hypoechoic lobulation of cortex (arrows) due to metastatic
deposit (T). H = hilum.
|
|
View larger version (109K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 8D —42-year-old woman with invasive ductal carcinoma and
metastatic type 5 and 6 nodes. Sonogram (D), histopathologic photograph
(E), and diagram (F) show type 6 completely hypoechoic node with
no hilum owing to metastatic replacement (T).
|
|
View larger version (133K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 8E —42-year-old woman with invasive ductal carcinoma and
metastatic type 5 and 6 nodes. Sonogram (D), histopathologic photograph
(E), and diagram (F) show type 6 completely hypoechoic node with
no hilum owing to metastatic replacement (T).
|
|
View larger version (5K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 8F —42-year-old woman with invasive ductal carcinoma and
metastatic type 5 and 6 nodes. Sonogram (D), histopathologic photograph
(E), and diagram (F) show type 6 completely hypoechoic node with
no hilum owing to metastatic replacement (T).
|
|
The long-axis size of nodes was measured in vitro with sonography.
Interobserver agreement in classifying nodes into different types and in
classifying them as benign or malignant was calculated from tabulated data.
Sonographic classification was judged against the histologic reference by
measurement of sensitivity, specificity, positive predictive value (PPV),
negative predictive value (NPV), overall accuracy, and positive and negative
likelihood ratios.
Results
A total of 186 pieces of tissue were scanned. Both reviewers
sonographically classified 15 pieces no node. At histologic examination, all
15 of these pieces were confirmed to be formless pieces of fat and were
therefore excluded from analysis. A total of 171 lymph nodes from 19 axillae
in 18 patients (age range, 31–71 years) were evaluated with in vitro
sonography. Seventeen of the 19 axillae, including the axilla of one patient
with bilateral cancer, contained invasive ductal carcinoma; two axillae
contained invasive lobular carcinoma, and 10 contained ductal carcinoma in
situ and invasive ductal carcinoma. Although neoadjuvant chemotherapy had been
administered to nine of 18 patients, their axillae were included because nodes
had positive findings in each of these cases.
Among 171 lymph nodes evaluated, 20–22% of the nodes were classified
type 1, 2, 4, or 5. Eight percent of the nodes were categorized type 3 and 7%
type 6. Interobserver agreement was 77% for nodal morphologic features (types
1–6) and 88% for characterization of a node as benign or malignant.
Interobserver agreement for classifying nodes was type 1, 71%; type 2, 89%;
type 3, 28%; type 4, 58%; type 5, 83%; and type 6, 83%. The total number of
histologically benign nodes was 149 (87.1%) and of malignant nodes was 22
(12.9%). Four nodes with false-negative findings had been classified type 4
and one node was classified type 3 at sonography. Thirty nodes with
false-positive findings were either normal or reactive hyperplasia. The
sensitivity, specificity, PPV, NPV, and overall accuracy were 77%, 80%, 36%,
96%, and 80%. The 95% CIs are shown in
Table 1. The positive
likelihood ratio was 3.84 (95% CI, 2.55–4.89), and the negative
likelihood ratio was 0.28 (95% CI, 0.13–0.54).
Table 2 summarizes the
performance of the node classification in the detection of metastasis.
The sizes of the various node types are shown in
Table 3. Using the Tukey
honestly significant difference procedure to perform all pairwise comparisons,
we found type 6 nodes to be significantly smaller than type 4 (p <
0.001) and type 2 (p < 0.001) nodes and marginally smaller than
type 5 nodes (p = 0.07) but not different in size from types 1 and 3
nodes. All type 6 nodes measured 1 cm or smaller. Among type 5 nodes, 21 of 35
measured 1 cm or smaller. Type 5 nodes with positive pathologic results were
significantly larger than type 5 nodes with negative pathologic results
(p = 0.04). Type 6 nodes with positive results were not statistically
different from type 6 nodes with negative results (p = 0.09), but the
number of data points (12) was small.
Figures 1,
2,
3 show the
sonographic–anatomic correlation of lymph nodes. Figures
4A,
4B,
4C,
5A,
5B,
5C,
6A,
6B,
6C,
7A,
7B,
7C,
8A,
8B,
8C,
8D,
8E,
8F show the pathologic and
sonographic findings for each lymph node type. In one of two cases of lobular
cancer, two nodes were classified type 5; both had positive pathologic
results. In the other case, all nodes were classified type 1 or 2, and all
nodes were found to be normal at histologic examination. The patient was
disease free 5 years after the examination.
Discussion
The historical reference standard for determining the presence of nodal
metastasis has been surgical axillary lymph node dissection, an invasive
procedure with high morbidity
[18]. The need for a less
invasive predictor of metastatic disease has led to increased use of
sonography, which has good resolution for detection of small nodes. Although
the long-axis cross-sectional measurement of nodes in our sample varied from
0.2 to 3.8 cm, we were able to clearly identify subcentimeter nodes of all
types (Table 3), affirming the
leading role of sonography in the detection of regional nodes in breast
cancer. Because all of the type 6 nodes and most of the type 5 nodes measured
1 cm or smaller, it appears that identifying these nodes was relatively easy
because of their fully or partially hypoechoic texture. Conspicuity of nodes,
therefore, appears to depend more on echogenicity than on size.
Because the mean sizes of benign and malignant nodes were similar, a
cortical morphologic feature, hypoechoic cortex, appears to be of greater
importance than node size. Type 6 nodes, although a smaller sample
(Table 3), appeared
significantly smaller than type 2, 4, and 5 nodes but not different from type
1 and 3 nodes. Although this finding would appear to suggest that type 6
metastatic nodes were inherently smaller nodes, such an inference would
require a much larger sample of these nodes. Although type 5 nodes with
positive pathologic results were statistically larger than type 5 nodes with
negative results, such a conclusion cannot be definitively made about type 6
nodes, possibly because there were too few of this type of node. We measured
only largest nodal diameter to judge conspicuity and did not measure
cross-sectional size ratios.
Although the sonographic criteria for nodal malignancy have centered on
node shape and overall echogenicity with some attention to vascularity
[19], little has been written
on cortical morphologic features. Metastatic cells in lymphatic fluid arrive
in the marginal sinus at the periphery of a lymph node through one or more
afferent lymphatic channels (Fig.
1). Lymph then filters from the marginal sinus through the cortex
and paracortex, which contain lymphocytes and phagocytes, toward the hilum
[12]. Metastatic deposits are
arrested by these cells in the periphery of a node
[12], causing enlargement of
the cortex, which can be localized or eccentric. This cortical bulge,
therefore, often may precede generalized enlargement of the cortex and
eventual replacement of the entire node. This type 5 appearance of smaller
nodes can pass unrecognized before turning into the type 6 appearance.
Besides a round node, the most commonly quoted criterion for malignancy has
been a hypoechoic node [5].
Although total replacement or displacement of nodal hilum makes a hypoechoic
lymph node suspicious, this criterion probably represents one extreme of a
spectrum of sonographic appearances of lymph nodes in metastatic disease. It
does not address earlier stages of metastatic disease, in which nodal cortex
(hypoechoic) and nodal hilum (hyperechoic) are coexisting areas of variable
shape. The unstimulated cortex of a benign node is extremely thin and almost
unrecognizable on sonograms (Fig.
4A), but the hilum is hyperechoic because it contains blood
vessels, fat, cords of lymphatic tissue, and a medullary sinus. Progressive
enlargement of a focal metastatic deposit, which might have started as a focal
or eccentric cortical bulge, eventually results in complete replacement or
displacement of the hilum, leading to the commonly described hypoechoic
node.
Assuming type 5 and 6 nodes (low PPV) to be malignant and type 1–4
(high NPV) nodes to be benign is rational because the assumption is based on
the anatomic features of the nodes and because false-negative nodes would
still be diagnosed correctly in the clinical setting owing to sentinel lymph
node mapping. Type 1 and 2 nodes (Figs.
4A,
4B,
4C and
5A,
5B,
5C) invariably can be regarded
as benign (Table 2) because
negligible cortical lymphatic tissue is visible. For type 3 nodes, in which
cortical thickness increases to more than 3 mm in a symmetric manner (Fig.
6A,
6B,
6C), the NPV is high (93%).
These nodes usually are reactive. The relatively low interobserver agreement
(28%) in classifying type 3 nodes might have been partly due to the small size
of the sample of these nodes compared with other types and to possible
judgment error regarding cortical thickness and misclassification as a type 2
appearance. This misclassification between types 3 and 2 appears to have
little clinical significance because both types were considered benign for
this study. In type 4 nodes (Fig.
7A,
7B,
7C), the presence of
generalized lobulation is parallel to the hilar contour.
Single false-negative nodes were found in five axillae. The size varied
from 0.6 to 1.8 cm, a type 3 node with micrometastasis being the smallest.
Although type 4 nodes are benign in the overwhelming majority of cases, four
false-negative type 4 nodes were found. Classifying type 4 nodes as malignant
would increase the sensitivity of this classification but would result in a
large number of negative findings at sonographically guided FNA. In the
clinical situation, without sonographically guided FNA, this small number of
false-negative nodes would be detected during sentinel lymph node mapping with
no compromise of patient care. Overall, type 1–3 nodes can be regarded
as benign and type 4 as likely benign.
Type 5 nodes (Figs. 8A,
8B,
8C) present a diagnostic
challenge because our pathophysiologic hypothesis of eccentric lobulation due
to localized metastatic deposition in the nodal cortex does not exclude the
possibility of normal-variant lobulation or inflammatory reactive changes in a
node. Therefore, although there were a large number of false-positive findings
(PPV, 29%) due to reactive changes, the implication is that type 5 nodes
should be sampled with sonographically guided FNA. In the type 5 appearance,
sonographically guided FNA should be directed into the focal cortical
lobulation. If the result of sonographically guided FNA is positive, the
patient may be spared a sentinel lymph node mapping procedure.
Type 6 nodes (Figs. 8D,
8E,
8F) are more clearly abnormal
than type 5 nodes (PPV, 58%), whether the lesion is metastasis, reactive
change, or another disease process such as lymphoma. Many false-positive
findings are made, but if the result of sonographically guided FNA is
negative, sentinel node mapping would be performed. Nineteen of 30
false-positive results occurred in only three axillae. These nodes were mainly
reactive, but one axilla had seven true-positive and five false-positive nodal
findings.
Lobular carcinoma has been notoriously difficult to diagnose
sonographically and histopathologically
[1,
10,
18], but we included both of
our cases because 5-year clinical follow-up showed our sonographic and
pathologic results to be concordant with the clinical state of the patient.
Other pitfalls and confounding appearances on sonography are as follows:
Relatively hypoechoic areas in the hilum in many nodes may be due to the
presence of fewer echogenic structures, such as blood vessels, and more
uniform fat replacement (Figs.
4A and
5A) but should not be mistaken
for hypoechoic cortex because of the intervening transitional zone of
paracortex. Hypoechoic nodal cortex is thinner on sonography than at
histologic examination (types 1 and 2) (Figs.
4A,
4B,
4C and
5A,
5B,
5C) because deeper cortical
layers (paracortex) are infiltrated with fat cells extending from the hilum.
Sonography requires careful technique (even in the clinical situation); care
must be taken to avoid tangential scans on which nodes can be
misclassified.
The relatively low prevalence of malignancy and the high NPV in our sample
might have resulted from exclusion of patients with nodes that were known to
have positive results from preoperative sonographically guided FNA. In
addition, three axillae had positive sentinel nodes (not included for
sonographic–pathologic correlation) and negative nodes throughout the
rest of the axilla. This situation was unavoidable because clinical care
dictated that we exclude sentinel nodes to avoid delay of diagnosis while the
patient was under anesthesia. However, because our objective was mainly
sonographic–pathologic correlation of cortical morphologic features, the
data can be interpreted with useful clinical inferences.
Inclusion of patients treated with neoadjuvant chemotherapy was possible in
our sample because the treatment response was partial, and metastatic nodes
were found in all of these patients. The relatively small number of nodes
makes verification with a larger sample desirable; this process will be
logistically difficult, even in academic institutions. Translation of this
classification into clinical practice remains to be proved. At this time we do
not advocate active classification of nodes in clinical practice. Our
objective was to propose a classification based on cortical morphologic
features rather than total node size and echogenicity so that it may serve as
a guide to performance of sonographically guided FNA. False-negative findings,
usually in type 4 nodes, resulting from use of our classification in a
clinical setting would be routinely addressed with sentinel lymph node
mapping, and patient care would not be compromised.
The results of in vitro sonography of dissected axillary nodes from
patients with breast cancer suggest that nodes can be classified on the basis
of cortical morphologic features for prediction of the occurrence of
metastatic disease. The presence of an asymmetric focal hypoechoic node with
cortical lobulation (type 5) or a completely hypoechoic node (type 6) should
serve as a guideline for universal performance of FNA in the preoperative
staging of breast cancer. This classification may be a useful clinical
guideline and should be further proved with in vivo studies.
Acknowledgments
The illustrations were prepared by David Bier and the manuscript by Elaine
Nitschke. We acknowledge contributions and participation of Isabelle
Bedrosian, S. E. Singletary, Barry Feig, Merrick I. Ross, Frederick C. Ames,
Henry M. Kuerer, Funda Meric-Bernstam, Gildy Babiera, and Val Johnson in the
study.
References
- Fisher B, Bauer M, Wickerham DL, Redmond CK, Fisher ER. Relation of
number of positive axillary nodes to the prognosis of patients with primary
breast cancer: NSABP update. Cancer 1983;28
: 282–291
- Carter CL, Allen C, Henson DE. Relation of tumour size, lymph node
status, and survival in 24,740 breast cancer cases.
Cancer 1989; 63:181
–187[CrossRef][Medline]
- Pamilo M, Soiva M, Lavast E-M. Real-time ultrasound, axillary
mammography, and clinical examination in the detection of axillary lymph node
metastases in breast cancer patients. J Ultrasound Med1989; 8:115
–120[Abstract]
- Lam WWM, Yang WT, Chan YL, Stewart IET, Metreweli C, King W.
Detection of axillary lymph node metastases in breast carcinoma by
technetium-99m sestamibi breast scintigraphy, ultrasound and conventional
mammography. Eur J Nucl Med 1996;23
: 498–503[Medline]
- Alvarez S, Añorbe E, Alcorta P, López F, Alonso I,
Cortés J. Role of sonography in the diagnosis of axillary lymph node
metastases in breast cancer: a systematic review. AJR2006; 186:1342
–1348[Abstract/Free Full Text]
- Krishnamurthy S, Sneige N, Bedi DG, et al. Role of
ultrasound-guided fine-needle aspiration of indeterminate and suspicious
axillary lymph nodes in the initial staging of breast carcinoma.
Cancer 2002; 95:982
–988[CrossRef][Medline]
- Yang WT, Ahuja A, Tang A, Suen M, King W, Metreweli C. High
resolution sonographic detection of axillary lymph node metastases in breast
cancer. J Ultrasound Med 1996;16
: 241–246
- Verbanck J, Vandewiele I, De Winter H, Tytgat J, Van Aelst F,
Tanghe W. Value of axillary ultrasonography and sonographically guided
puncture of axillary nodes: a prospective study in 144 consecutive patients.
J Clin Ultrasound 1997;25
: 53–56[CrossRef][Medline]
- Vaidya JS, Vyas JJ, Thakur MH, Khandelwal KC, Mittra I. Role of
ultrasonography to detect axillary node involvement in operable breast cancer.
Eur J Surg Oncol 1996;22
: 140–143[CrossRef][Medline]
- Bedrosian I, Bedi D, Kuerer HM, et al. Impact of
clinicopathological factors on sensitivity of axillary ultrasonography in the
detection of axillary nodal metastases in patients with breast cancer.
Ann Surg Oncol 2003;10
:1025
–1030[CrossRef][Medline]
- Fornage BD. Interventional ultrasound of the breast. In: McGahan
JP, ed. Interventional ultrasound. Baltimore, MD:
Williams & Wilkins, 1990:71
–83
- van der Valk P, Meijer CJL. In: Sternberg SS, ed.
Histology for pathologists, 2nd ed. Philadelphia, PA:
Lippincott Raven, 1997:651
–673
- Feu J, Tresserra F, Fábregas R, et al. Metastatic breast
carcinoma in axillary lymph nodes: in vitro US detection.
Radiology 1997;205
: 831–835[Abstract/Free Full Text]
- Tateishi T, Machi J, Feleppa EJ, et al. In vitro B-mode
ultrasonographic criteria for diagnosing axillary lymph node metastasis of
breast cancer. J Ultrasound Med 1999;18
: 349–356[Abstract]
- Vassallo P, Edel G, Roos N, Naguib A, Peters PE. In-vitro
high-resolution ultrasonography of benign and malignant lymph nodes: a
sonographic-pathologic correlation. Invest Radiol1993; 28:698
–705[CrossRef][Medline]
- Rubaltelli L, Proto E, Salmaso R, Bortoletto P, Candiani F, Cagol
P. Sonography of abnormal lymph nodes in vitro: correlation of sonographic and
histologic findings. AJR 1990;155
:1241
–1244[Abstract/Free Full Text]
- Uematsu T, Sano M, Homma K. In vitro high-resolution helical CT of
small axillary lymph nodes in patients with breast cancer: correlation of CT
and histology. AJR 2001;176
:1069
–1074[Abstract/Free Full Text]
- Cady B. The need to reexamine axillary lymph node dissection in
invasive breast cancer (editorial). Cancer1994; 73:505
–508[CrossRef][Medline]
- Yang WT, Chang J, Metreweli C. Patients with breast cancer:
differences in color Doppler flow and gray-scale US features of benign and
malignant axillary lymph nodes. Radiology2000; 215:568
–573[Abstract/Free Full Text]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
|
|
 
N. Cho, W. K. Moon, W. Han, I. A. Park, J. Cho, and D.-Y. Noh
Preoperative Sonographic Classification of Axillary Lymph Nodes in Patients With Breast Cancer: Node-to-Node Correlation With Surgical Histology and Sentinel Node Biopsy Results
Am. J. Roentgenol.,
December 1, 2009;
193(6):
1731 - 1737.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
