Diagnostic Value of Fine-Needle Aspiration in Supraclavicular ...

Diagnostic Value of Fine-Needle
Aspiration in Supraclavicular
Lymphadenopathy:
A Study of 106 Patients and Review of
Literature
Joseph F. Nasuti, M.D., 1 Ravi Mehrotra, M.D., M.I.A.C., 2 * and
Prabodh K. Gupta, M.D., F.I.A.C. 1
The simplicity and safety of the fine-needle aspiration (FNA) make
it a useful procedure for the initial evaluation of supraclavicular
lymphadenopathy. We analyzed data (106 patients) in combination
with the data of three recently published studies (total 301 patients)
to make the observations more meaningful. The usefulness
of various ancillary techniques such as special histochemical
stains (SHC), immunohistochemistry (IHC), and flow cytometry
(FCM) in establishing specific diagnoses was also examined. Analysis
of the combined data from 407 patients helps to reinforce
certain trends observed by previous investigators such as the
marked predominance of metastatic carcinoma (63%) over lymphoid
malignancies (9%), reactive (19%) and infectious processes
(2%), and the tendency of abdominal and pelvic primary carcinomas
to metastasize to the left supraclavicular lymph node (SCLN)
(88%). In addition, the current study confirms utility of performing
IHC, FCM, and SHC on the aspirate material obtained from
supraclavicular lymph nodes, because in 19 of the 22 cases (86%),
these ancillary techniques provided supplemental diagnostic information
critical to patient management. Diagn. Cytopathol.
2001;25:351–355. © 2001 Wiley-Liss, Inc.
Key Words: supraclavicular lymph node; fine-needle aspiration;
immunohistochemistry; flow cytometry
Fine-needle aspiration (FNA) is a simple and safe procedure
that has proven to be accurate in the diagnosis of a variety
of disease processes. It can be performed in an outpatient or
inpatient setting and results are routinely available within 24
1 University of Pennsylvania Medical Center, Department of Pathology
and Laboratory Medicine, Cytopathology and Cytometry Section, Philadelphia,
Pennsylvania
2 Department of Pathology, Moti Lal Nehru Medical College, Allahabad,
India
This project initiated by Ravi Mehrotra, M.D., M.I.A.C.
*Correspondence to: Joseph F. Nasuti, M.D., University of Pennsylvania
Medical Center, Department of Pathology and Laboratory Medicine, Cytopathology
and Cytometry Section, 3400 Spruce Street/6 Founders Pavilion,
Philadelphia, Pennsylvania 19104. E-mail: jfnasuti@mail.med.upenn.edu
Received 15 February 2001; Accepted 7 June 2001
hours. A preliminary diagnosis is often obtained within a
few minutes of FNA when immediate staining and microscopic
examination is performed. The use of ancillary techniques
such as IHC (immunohistochemistry) and FCM
(flow cytometry) have greatly improved the diagnostic accuracy
of FNA in general and particularly in cases of
lymphadenopathy. 1–4
Supraclavicular lymph node (SCLN) (also known as the
sentinel lymph node), in many instances is the first sign of
an underlying malignancy in the thoracic cavity, abdominal
cavity, or the pelvic region. 5 The left SCLN (also called the
Virchow’s lymph node) has been known to be a common
site of distant metastasis in the spread of gastric cancer. 6
The easy accessibility of enlarged SCLNs offers an ideal
target for FNA by palpation alone. A variety of benign and
malignant conditions presenting as supraclavicular lymphadenopathy
and diagnosed on FNA have been described in
the recent literature. 7–15 To the best of our knowledge, no
study on FNA of the SCLN has been published that adequately
addresses the diagnostic utility of integrating FCM
and IHC along with cytomorphologic findings.
We reviewed our experience with FNA of the SCLN over
a 10-yr period to determine the spectrum of diseases processes.
The observations have been compared with previously
published studies.
Materials and Methods
The medical records of all patients who underwent FNA of the
SCLN between January 1, 1991, and December 31, 2000, (10
yr) at the University of Pennsylvania Medical Center were
evaluated for age; gender; results of FNA, IHC, and FCM;
histologic diagnosis; and clinical follow-up. Aspirates were
procured by using a 23- or 25-gauge, 1- and 0.5-inch needle
attached to a 10-ml plastic syringe. Generally, one to three
passes were performed; additional aspirates were made to
© 2001 WILEY-LISS, INC. Diagnostic Cytopathology, Vol 25, No 6 351
DOI 10.1002/dc.10002

NASUTI ET AL.
Fig. C-1. A: FNA of tuberculosis, note the epithelioid granuloma in a background of acute inflammation and necrotic debris (Papanicolaou stained, 95%
ethanol-fixed direct smear, 200). B: Note the magenta-stained bacilli scattered within the cytoplasm of histocytes and in the background of the same FNA
material obtained from the case depicted in Figure 2A (acid-fast stain performed on 95% ethanol-fixed direct smear, 1,000).
Fig. C-2. A: Fine-needle aspiration (FNA) of metastatic prostate carcinoma containing pleomorphic epithelial tumor cells characterized by irregular
vesicular nuclei with prominent nucleoli (Papanicolaou stained, 95% ethanol-fixed direct smear, 400). B: Note the cytoplasmic staining of prostatespecific
antigen (PSA) in tumor cells from the cell block FNA material obtained from same case depicted in Figure 1A (PSA antigen-antibody reaction
performed on cell block processed material with diaminobenzidine chromagen, 800).
increase the harvest of cells for possible ancillary studies (FCM
and IHC). Aspirated material was smeared onto glass slides,
with half the smears immediately fixed in 95% alcohol for
Papanicolaou staining and the other half allowed to air-dry for
Diff-Quick stain (Becton-Dickinson, San Jose, CA). The residual
material from each pass was rinsed in Normosol solution
for Millipore filter preparation, cell block, and possible
ancillary studies (FCM and IHC).
352 Diagnostic Cytopathology, Vol 25, No 6

ASPIRATION IN SUPRACLAVICULAR LYMPHADENOPATHY
On-site cytomorphologic evaluation was performed by
examination of air-dried, Diff-Quik stained (Becton-Dickinson)
smears to provide preliminary diagnostic information
to the clinicians and to determine the need based on cellular
adequacy for additional passes to optimize diagnostic accuracy
or for possible ancillary studies. The results of the final
cytology diagnosis from each case were based on the interpretation
of all the cytology material obtained and processed
as described above along with all available clinical information
and data obtained from any ancillary studies performed
(FCM and IHC). The results were categorized as
benign/reactive, malignant ,or nondiagnostic and compared
with histologic results when available.
Results
Of the 17,937 fine-needle aspirations performed over a
10-yr period (January 1, 1991, to December 31, 2000) 106
patients (108 cases, 0.6%) underwent the diagnostic procedure
for the evaluation of an enlarged supraclavicular lymph
node with two patients being aspirated twice due to a
nondiagnostic aspirates in the first attempt. The average age
of the patients was 59 yr with a range of 26-90 yr; 49 were
male and 56 female. The average size of the supraclavicular
lymph nodes aspirated was 2.3 cm with a range of 1.5 to 5
cm. A malignant tumor was diagnosed on FNA in 77 (73%)
patients. Follow-up excisional biopsies were available in 19
cases (18%) with no false-positive or -negative results.
Forty-four patients (57%) with a malignant diagnosis on
FNA had a prior tissue diagnosis of cancer, whereas 33
patients (43%) were newly diagnosed with malignancy.
Five patients (5%) presented with an abnormal lymph node
that was classified as benign reactive hyperplasia by FNA.
Two patients were diagnosed with tuberculosis based on the
cytomorphologic findings of a granulomatous inflammation
and the results of a Kinyoun stain that confirmed the presence
of acid-fast bacilli in the aspirate material obtained
(Fig. C-1). Two (2%) of the 108 cases were nondiagnostic
on FNA, both of which were subsequently diagnosed with
malignancy: one on follow-up FNA alone (metastatic adenocarcinoma,
lung primary), and the other on repeat FNA
and subsequent surgical excision (metastatic teratoma, mediastinal
primary).
Of the 77 patients diagnosed with malignancy, 69(90%)
were nonlymphoid (66 metastatic carcinomas, 1 melanoma,
1 teratoma, and 1 liposarcoma) and 8 (10%) were lymphoid
(5 non-Hodgkin’s B-cell lymphomas and 3 Hodgkin’s lymphomas).
These lesions were distributed nearly equally 49%
(37 cases) and 51% (39 cases) between the right and left
sides, respectively. A similar distribution was observed
among supradiaphragmatic tumors: 15 lung (7 right, 8 left)
and 10 breast (5 right, 5 left) carcinomas. The 10 infradiaphragmatic
primary malignancies (3 gastrointestinal, 2
bladder, and 1 ovarian, 1 cervical, 1 endometrial, 1 renal, 1
Table I.
Type of study
performed
Summary of Ancillary Studies a
Number of cases
performed
Number
of cases
positive
Number of cases
diagnostically
helpful results
IHC 12 10 11
Cytokeratins 8 6 7
PSA, PSAP 3 1 3
EMA, vimentin 2 2 2
Neuroendocrine
markers 4 2 3
Germ cell markers 2 0 2
CEA 1 0 0
LCA 2 1 2
ER, PR 2 0 0
SHC 8 5 5
Mucicarmine 6 3 3
Kinyoun 2 2 2
FCM 5 5 5
B-cell markers 3 3 3
T-cell markers 2 2 2
Monoclonality 3 3 3
a IHC, immunohistochemistry; cytokeratins, AE1/AE3, CAM 5.2, PAN
CK; PSA, prostate specific antigen; PSAP, prostatic acid phosphatase;
EMA, epithelial membrane antigen; neuroendocrine markers, neuron specific
enolase, synaptophysin, chromogranin; germ cell markers, placental
alkaline phosphatase, alpha-fetal protein, human chorionic gonotropin;
CEA, carcino embryonic antigen; LCA, leukocyte common antigen; ER,
estrogen receptors; PR, progesterone receptors; SHC, special histochemical
stains; FCM, flow cytometry.
prostate) did, however, demonstrate a preference for the left
SCLN with 8 of the 10 (80%) metastasizing to that side.
Ancillary studies including special histochemical stains
(8), IHC (12), and FCM (5) were performed on the aspirate
material obtained in 22 (21%) cases, with special histochemical
stains and IHC used jointly in three cases (Table
I). IHC was useful in further characterizing tumors in 10 of
the 12 cases in which it was used, including confirming the
presence of metastatic renal cell carcinoma in two patients
with a known primary and in the diagnosis of metastatic
prostate carcinoma in patient with no known primary (Fig.
C-2). In addition to usefulness the acid fast stain in the
diagnosis of tuberculosis, special histochemical stain mucicarmine
was also helpful in further typing poorly differentiated
carcinomas as adenocarcinomas in three of six (50%)
cases. In one of these cases, mucicarmine stain was helpful
in confirming a diagnosis of a metastatic poorly differentiated
rectal adenocarcinoma.
Flow cytometry was performed on the aspirate material
of five patients presenting with supraclavicular lymphadenopathy
suspected of non-Hodgkin’s lymphoma. These include
two patients with reactive hyperplasia in whom FCM
revealed a mixed population of predominately small T-
lymphocytes with no evidence of monoclonality. Three
patients subsequently diagnosed with lymphoma also had
FCM performed, which revealed a monotypic B-cell population,
two of which were CD19 and CD20 positive, CD5
negative, and lambda restricted. The remaining case was
CD19, CD20, CD23, CD10 positive and kappa restricted.
Diagnostic Cytopathology, Vol 25, No 6 353

NASUTI ET AL.
Table II.
Comparison With Other Studies a
Authors Cervin (1995) Carson (1996) McHenry (1999) Present (2001) Total
Number of patients 149 100 52 106 407
Malignancies 96 (64%) 83 (83%) 42 (81%) 77 (73%) 298 (73%)
Metastatic carcinomas 85 70 35 66 256
Supradiaphragmatic
primary T 51, L 26, R 25 T 38, L 21, R 17 T 20, L 10, R 10 T 25, L 12, R 13 T 134, L 69, R 65
Infradiaphragmatic
primary T 25, L 22, R 3 T 18, L 15, R 3 T 11, L 11, R 0 T 10, L 8, R 2 T 64, L 56, R 8
Lymphoid malignancies 10 11 9 8 38
Others malignancies 1 2 2 3 8
Reactive/benign 32 15 4 25 76
Infectious 8 — — 2 10
Nondiagnostic 13 2 6 2 23
a T, total; L, left supraclavicular lymph node; R, right supraclavicular node.
Discussion
Several studies, including our own, have found that a large
percentage of patients presenting with persistent supraclavicular
lymphadenopathy were subsequently diagnosed
with malignancy based on tissue studies (58–83%). 7–12,15
Of these, the vast majority (65–90%) are metastatic carcinomas
from primary tumors originating from visceral sites
inferior to the supraclavicular nodal groups. Viacava and
Pack in 1944 reviewed 4,365 patients with primary visceral
cancers and found that a disproportionate number of abdominal
and pelvic tumors metastasized to the left SCLN. 7
More recent FNA studies involving a combined 306 patients
reported that 83–100% of known abdominal and pelvic
primary carcinomas metastasized to the left SCLN (Table
II). 8–10 These same studies found no such preference for
laterality in SCLN metastasis from thorax, head, and neck
malignancies. The results of the current study are similar,
with 80% of infradiaphragmatic tumors metastasizing to the
left SCLN and a virtually even distribution for supradiaphragmatic
primary tumors (right 48%, left 52%).
On-site evaluation of FNA Diff-Quik stained specimen,
we believe is extremely valuable in not only providing an
accurate on the spot interpretation in a large majority of
cases (4,829 of 5,872, 82% unpublished data) but also help
triage specimens for ancillary studies. The utility of ancillary
techniques such as IHC, flow cytometry, and special
histochemical stains on SCLN FNA pertaining to the
amount of additional relevant clinical information provided
has not, to our knowledge, been addressed in previous
reports. In the current study, these techniques were used in
21% (22 of 107) of the cases; these methods led to useful
diagnostic information in 19 of 22 cases (86%) including
detection of acid-fast bacilli in a background of granulomatous
inflammation; mucin in poorly differentiated carcinoma;
the presence of PSA, LCA, EMA, and vimentin in
metastatic cancer cells, which either confirmed a metastasis
from a known primary or helped to establish the tissue of
origin in patients who initially presented with metastatic
disease. As was the case in the previously published reports
(7–13%) lymphoid malignancy represented a minority of
the tumors in the current study (9%); however, the use of
FCM on the FNA material obtained was found to be extremely
helpful in ruling out lymphoma in two patients with
reactive lymphadenopathy and in the detection and characterization
of a monoclonal population of lymphoma cells in
three patients.
The data obtained in the current study and in recently
published reports confirm the diagnostic utility of FNA in
the evaluation of patients presenting with supraclavicular
lymphadenopathy. Also, the combined data from this and
the prior studies (407 patients) help to reinforce certain
trends observed by previous investigators such as the
marked predominance of metastatic carcinoma over lymphoid
malignancies, reactive and infectious processes along
with the tendency of abdominal and pelvic primary carcinomas
to metastasize to the left SCLN. 8–10 In addition, the
current observations support the use of IHC, FCM, and
special histochemical stains on the aspirate material obtained
from SCLN. These ancillary techniques have shown
the potential in some cases to provide additional clinical
information critical to patient management. This potential
may be increased in the near future due to the development
of new immunohistochemical antibodies to tissue specific
markers such as uroplakin (urothelium) detectable in metastatic
transitional cell carcinoma. 16
References
1. Nasuti JF, Yu G, Boudusquie A, Gupta PK. Diagnostic value of lymph
node fine-needle aspiration cytology: an institutional experience of 387
cases observed over a 5-year period. Cytopathology 2000;11:18–31.
2. Cafferty LL, Katz RL, Ordenz NG. Fine-needle aspiration diagnosis of
intraabdominal and retroperitoneal lymphomas by a morphologic and
immunocytochemical approach. Cancer 1990;65:72–77.
3. Robins DB, Katz RL, Swan F. Immunophenotyping of lymphoma by
fine-needle aspiration: a comparison study of cytospin preparations
and flow cytometry. Am J Clin Pathol 1994;101:569–576.
4. Sneiger N, Kekmezian R, El-Naffer A, Manning J. Cytomorphological,
immunocytochemical and nucleic acid flow cytometric study of 50
lymph nodes by fine-needle aspiration. Cancer 1991;67:1003–1007.
354 Diagnostic Cytopathology, Vol 25, No 6

ASPIRATION IN SUPRACLAVICULAR LYMPHADENOPATHY
5. Moore KL. The thorax. In: Moore KL, editor. Clinically oriented
anatomy, 2nd ed. Baltimore: William and Wilkins; 1985. p 49–148.
6. Morgenstern L. The Virchow-Troisier node: a historical note. Am J
Surg 1979;138:703.
7. Viacava EP, Pack GT. Significance of supraclavicular signal node in
patients with abdominal and thoracic cancer. Arch Surg 1944;48:109–
119.
8. Cervin JR, Silverman JF, Loggie BW, Geisinger KR. Virchow’s node
revisited: analysis with clinicopathologic correlation of 152 fine-needle
aspiration biopsies of supraclavicular nodes. Arch Pathol Lab Med
1995;119:727–730.
9. Carson HJ, Candel AG, Gattuso P, Castelli MJ. Fine-needle aspiration
of supraclavicular lymph nodes. Diagn Cytopathol 1996;14:
216–220.
10. McHenry CR, Cooney MW, Slusarczyk SJ, Khiyami A. Supraclavicular
lymphadenopathy: the spectrum of pathology and evaluation by
fine-needle aspiration biopsy. Am Surg 1999;65:742–746.
11. McMenamin PG, Anderson JD, Baker JJ. Prostatic carcinoma presenting
with neck metastasis. J Otolaryngol 1989;18:119–124.
12. Yoa ZH, Wu AR. Supraclavicular lymph node metastasis from carcinoma
of the uterine cervix after radiotherapy: analysis of 219 patients.
Chung-Hua Chung Liu Tsa Chih 1988;10:230–232.
13. Ereno C, Lopez JI, Elizalde JM, Ibanez T, Fernandez-Larriano A,
Toledo JD. A case of Whipples’s disease presenting as supraclavicular
lymphadenopathy. A case report. APMIS 1993;101:865–868.
14. Pithie AD, Chicksen B. Fine-needle extrathoracic lymph node aspiration
in HIV associated sputum negative tuberculosis. Lancet 1992;340:
1504–1505.
15. Khorsand J, Katz RL, Savaraj N. Malignant carcinoid of pancreas: a
cytologic, ultrastructural and immunocytochemical study of a case
diagnosed by fine-needle aspiration of a supraclavicular node metastasis.
Diagn Cytopathol 1987;3:222–227.
16. Xu X, Sun TT, Gupta PK, Nasuti JF. Uroplakins as a marker for typing
metastatic transitional cell carcinoma on fine-needle aspiration specimens.
Cancer Cytopathol 2001;83:216–221.
Diagnostic Cytopathology, Vol 25, No 6 355