Ultrasound-Guided Fine-Needle Biopsy of Neck Nodes Robert J. Baatenburg de Jong, MD, PhD; Robert J. Rongen, MD, PhD; Carel D. A. Hans van Overhagen, MD; Johan S. Lam\l=e'\ris,MD, PhD; Paul Knegt, MD, PhD
\s=b\ The assessment of nodal involvement in patients with squamous cell carcinoma of the head and neck is still a major diag-
nostic problem. Although the sensitivity of imaging techniques for detection of neck nodes is gradually improving, the specificity for metastases remains low. Cytologic examination could, theoretically, supply additive information. Computed tomographic\p=n-\and magnetic resonance\p=n-\guided aspiration techniques have been described, but these were not efficacious and laborious. In 1984, we developed a technique for ultrasound-guided (UG) fine\x=req-\ needle aspiration biopsy (FNAB). This technique is described herein, and the value of UGFNAB is compared with conventional FNAB. All statistical characteristics of UGFNAB appeared to be superior to conventional FNAB (sensitivity, 98% vs 88%; specificity, 95% vs 82%; positive predictive value, 98% vs 93%; negative predictive value, 95% vs 74%; and accuracy, 97% vs 87%). Furthermore, UGFNAB was characterized by less nondiagnostic aspirations. It is concluded that UGFNAB is a reliable technique for differentiation between benign nodes and cervical lymph node metastases and it may, therefore, contribute to a more accurate assessment of the neck in squamous cell carcinoma of the head and neck. (Arch Otolaryngol Head Neck Surg.
In patientsof with the head
squamous cell carci¬
noma
and neck, cyto¬
logie examination of neck nodes is a frequently used method to distinguish Accepted for publication December 20,1990. From the Departments of Otorhinolaryngology (Drs Baatenburg de Jong, Verwoerd, and Knegt) and Radiology (Drs van Overhagen and Lam\l=e'\ris), Dijkzigt University Hospital, Rotterdam; and Department of Radiology, Canisius-Wilhelmina Hospital, Nijmegen (Dr Rongen), the Netherlands.
Reprint requests to Department of Otorhinolaryngology, Dijkzigt University Hospital, Dr. Molenwaterplein 40, 3015GD Rotterdam, the Netherlands (Dr Baatenburg de Jong).
Verwoerd, MD, PhD;
between benign and malignant dis¬ ease. Frable and Frable1 note a 95% sensitivity for the presence of tumor in palpable cervical nodes and a 98% specificity for the absence of malig¬ nancy. These and other authors25 re¬ port an accuracy of conventional fineneedle aspiration biopsy (FNAB) vary¬ ing from 80% to 98%. With the introduction of small-part transducers for ultrasound equipment, many nonpalpable neck nodes are visualized68 (Fig 1). To allow differen¬ tiation between benign nodes and cer¬ vical lymph node métastases, these le¬ sions require cytologie evaluation by FNAB, which is necessarily performed under ultrasound guidance. In this ar¬ ticle, the technique of ultrasoundguided FNAB (UGFNAB) is described, and the results of cytologie examina¬ tion of material obtained by UGFNAB are compared with cytologie examina¬ tion of material from palpable nodes obtained by FNAB. Histopathologic examination of subsequently per¬ formed radical neck dissection was the gold standard of this study. MATERIALS AND METHODS Materials
1991;117:402-404)
Between January 1980 and January 1988, 143 cervical nodes of patients with squamous cell carcinoma of the upper aerodigestive tract were examined cytologically prior to (radical) neck dissection. From 1980 until 1985, FNABs only were per¬ formed (36 cases); and from 1985 onward, both FNABs (31 cases) and UGFNABs (76
cases)
were
Aspiration
employed.
FNAB Technique The well-known FNAB technique de¬ scribed by Frable and Frable' was applied, using the Cameco syringe-pistol with a 20-mL disposable syringe and a 0.6-mm needle. The FNABs were performed by a
Downloaded From: http://archotol.jamanetwork.com/ by a University of Pennsylvania User on 06/15/2015
head and neck
pathologist.
oncologist,
or
by
a
cyto¬
UGFNAB Technique Lymph nodes were visualized by ultra¬ sound examination with small-part trans¬ ducers. The examination was performed by
investigators (R.J.R., H.vO., J.S.L., or R.J.B.dJ.). During this study, the Aloka
four
SSD-650 with UST 5-MHz and 7.5-MHz transducers (Aloka Co Ltd, Japan) and Philips SDR 1500 with a 7.5-MHz trans¬ ducer were employed. Again, a 0.6-mm nee¬ dle mounted on a needle-holder (Cameco) was used; for very small and/or mobile le¬ sions, a butterfly needle was used instead. The latter needle allows more subtle ma¬ neuvering and is, therefore, a valuable al¬ ternative for the needle-holder. Once a mass is visualized, it is centered on the monitor, by placing it right under the middle of the transducer (Fig 2). When the lesion is displayed in this way, it is, when necessary, fixed to surrounding tissues by applying gentle pressure with the trans¬ ducer. While the sonographer provides op¬ timal imaging, a coworker introduces the needle into the overlying skin. Depending on the depth of the structure to be exam¬ ined, the angle between the needle and the skin can be varied (Fig 3). Once the tip of the needle is inside the lesion, it is recog¬ nized as a bright, echogenic structure and the plunger of the syringe is retracted, cre¬ ating a negative pressure in the needle lu¬ men. While a constant vacuum is main¬ tained, the needle is moved back and forth under ultrasound control, and the lesion is sampled in different areas (Fig 4). Subse¬ quently, the plunger is slowly released and the needle withdrawn.
Smear Preparation Thin smears from the aspirated material were air dried and stained with May-Grünwald-Giemsa staining methods.
Cytologie Examination All smears were examined by an experi¬ enced cytopathologist. Samples were clas-
Fig 1.—Section parallel to the sternocleido¬ mastoid muscle (scm) showing a small, nonpalpable node (n) between the sternocleido¬ mastoid muscle and the jugular vein (jv). Cyto¬ logie examination of material obtained by ultrasound-guided fine-needle aspiration bi¬
Fig 3.—The angle between the needle and the skin must be adjusted to the depth of the lesion.
the node.
opsy showed squamous cell carcinoma.
Table 1.—Results of FNAB and UGFNAB Compared With
Table 2.—Test Characteristics of FNAB, UGFNAB, and FNAB and
Histopathologic Examination"
UGFNAB*
Hist+ 38
FNAB+ FNAB-
Hist-
UGFNAB+ UGFNAB-
in 7
cases
55
1
1
18
in five
a
nondiagnostic aspirate. Histopathologic Examination The neck dissection specimens were pre¬
served in a formaldehyde solution and ex¬ amined by a senior pathologist. Macroscopically enlarged nodes were removed and sectioned for histopathologic examination. Furthermore, the specimens were sectioned at 3- to 4-mm intervals.
RESULTS
The results of FNAB performed in 67 cases were true-positive in 38 cases, true-negative in 14 cases, false-posi¬ tive in three cases, and false-negative
cases (Table 1); in seven cases, satisfactory diagnosis could be made because the aspirate did not contain lymphatic or epithelial cells (six smears) or was classified as Pap III (one smear). The test characteristics of FNAB were as follows: sensitivity of 88%, specificity of 82%, positive-pre¬ dictive value of 93%, negative-predic¬
FNAB
UGFNAB
(n 67)
(n 76)
88 82
98 95
93 74 87 10 72
98 95 97
=
14
Unsatisfactory for interpretation in 1 case *FNAB indicates fine-needle aspiration biopsy; UGFNAB, ultrasound-guided FNAB; Hist, histopatho¬ logic; plus sign, positive; and minus sign, negative.
sified according to the method of Papanicolaou (Pap). Retrospectively, Pap I and Pap II smears were considered benign (nega¬ tive) and Pap IV and Pap V smears were considered malignant (positive). A smear classified as Pap III was considered as being
FNAB and
3
5
Unsatisfactory for interpretation
Fig 2.—Once the lesion to be examined is vi¬ sualized, the middle of the transducer is placed superior to the center of the lesion.
Fig 4.—The visualization of the needle allows controlled sampling from different areas within
Sensitivity, Specificity,
% %
PPV, % NPV, % Accuracy, % No diagnosis, % Prevalence, %
=
UGFNAB
(n
=
143)
94 89 96 84 93
1
5
75
73
FNAB indicates fine-needle aspiration biopsy; UGFNAB, ultrasound-guided FNAB; PPV, positivepredictive value; and NPV, negative-predictive value.
no
tive value of 74 %, and accuracy of 87 %
(Table 2).
The results of UGFNAB performed in 76 cases can be specified as follows: 55 were true-positive, 18 were true-
negative, false-positive, and one was false-negative; in one sample only blood and muscle cells were ob¬ served and it was considered not satis¬ factory (Table 1). From these data, the following test characteristics were cal¬ culated: 98% sensitivity, 95% specific¬ ity, 98% positive-predictive value, 95% one was
negative-predictive value, and 97% ac¬ curacy (Table 2). The overall results of cytologie ex¬ amination (FNAB and UGFNAB)
were
characterized by 94% sensitivity,
89%
specificity,
96%
positive-predic¬
tive value, 84% negative-predictive value and 93% accuracy. In 5% of the aspirates, no diagnosis could be made
(Table 2).
There were no complications attrib¬ utable to the aspiration. COMMENT
With continuous
improvements in computed tomography, magnetic reso¬ nance imaging, and ultrasound, the sensitivity of imaging techniques for detection of neck nodes is gradually increasing. However, criteria that have been developed for differentia¬
tion between reactive nodes and mé¬ tastases are not satisfactory. The size criterion, for instance, is quite arbi¬ trary, eg, the lower the limit used as a cutoff between normal and malignant nodes, the higher the false-positive
Downloaded From: http://archotol.jamanetwork.com/ by a University of Pennsylvania User on 06/15/2015
Radiological criteria include the depiction of central necrosis. However, fatty nodal replacement may resemble rates.
central necrosis and this may occur in
postinflammatory and postirradiation nodes.9 Other radiological criteria like obliteration of fasciai planes and the visualization of contiguous nodes may occur in inflammatory disease as well.10 As it remains difficult to identify benign and metastatic disease on grounds of radiologie characteristics only, tissue diagnosis would be benefi¬ cial. However, it is well established
that open neck biopsy in cases of squamous cell carcinoma of the upper aerodigestive tract is detrimental."·12 In the head and neck region, FNAB provides an alternative to premature open biopsy of masses in the neck. The risk of tumor seeding in the tract of the needle is negligible.1315 No serious side effects of FNAB have been reported.216 Methodical use of an imaging mo¬ dality combined with cytological ex¬ amination has not yet been described. In this study, UGFNAB was charac¬ terized by less false-negative results than was FNAB. This is probably due to the continuous monitoring of the tip of the needle and the node during UGFNAB, which allows controlled sampling from different areas within a node, reducing the chance of leaving a part of a node unexamined. In view of the low number of false-negative re¬ sults of UGFNAB and the test param¬ eters of cytologie examination that are reported in the literature,151617 the value of a negative cytologie diagnosis should be reconsidered. Usually, pal¬ pable nodes are considered metastatically involved, irrespective of the out¬ come of cytologie examination. Our results, however, indicate once more that there is a high probability that nodes with a negative result on UGFNAB indeed are benign on histo¬ pathologic examination. In relevant literature, the value of a positive cytologie diagnosis of meta¬ static disease is undisputed. In our study, however, four false-positive re¬ sults were noted in 143 examined nodes. These nodes were classified as Pap IV (three cases; two by FNAB, one by UGFNAB) and Pap V (one case by FNAB). All nodes were present in pre-
viously irradiated necks and appeared to be necrotic on histopathologic ex¬ amination. There were no false-posi¬
excludes metastatic involvement with a high degree of reliability. (4) Our re¬ sults demonstrate once more that cy¬ tologie examination, when performed by a well-trained team of a cytopathol¬ ogist and a clinician or sonographer, is an accurate technique for differentia¬ tion between benign nodes and cervical lymph node métastases.
tive results from nodes that were not irradiated. Considering the fact that nodes as¬ pirated under ultrasound guidance
generally nonpalpable and usu¬ ally smaller than were the nodes aspi¬ rated by the conventional means, our
were
results indicate that UGFNAB is more efficient than is FNAB; in only one case of 76 UGFNABs no diagnosis could be made, while conventional FNAB yielded nonsatisfactory material in 10% of the aspirations. It is assumed that the sampling technique is respon¬ sible for the more advantageous re¬ sults of UGFNAB when compared with FNAB. The number of nondiagnostic aspirations may have also been influenced by the fact that patients were selected for this study on the ba¬ sis of the availability of the results of histopathologic examination, eg, when the result of cytologie examination is nondiagnostic, patients are less likely to be referred for the gold standard procedure. Therefore, in everyday clin¬ ical practice, the number of nondiag¬ nostic aspirations can be expected to be higher. The results of FNAB in this study seem to compare unfavorably with those of other authors.15 However, most of our false-negative results (four of five) were noted in the first years after introduction of FNAB in our institution. Furthermore, irradia¬ tion accounted for all false-positive results, whereas in other studies, irra¬ diated nodes were not included or pre¬ vious irradiation was not mentioned. Therefore, our results may not be as unfavorable as they seem. In summary, our results of study on patients with squamous cell carcinoma of the upper aerodigestive tract have led to the following conclusions: (1) The UGFNAB is (at least) as accu¬ rate as conventional FNAB, and per¬ mits cytologie examination of nonpal¬ pable lymph nodes. (2) False-positive results of cytologie examinations occur only in nodes that were previously ir¬ radiated. (3) High sensitivity, high specificity, and high negative-predic¬ tive value of UGFNAB indicate that a negative diagnosis by use of UGFNAB
References 1. Frable MA, Frable WJ. Fine needle aspiration biopsy revisited. Laryngoscope. 1982;92:1414\x=req-\ 1418. 2. Engzell U, Jacobsson PA, Stgurdson A, Zajicek J. Aspiration biopsy of metastatic carcinoma of lymph nodes in the neck. Acta Otolaryn-
gol. 1971;72:138-147. 3. Meyers DS, Templer J. Aspiration cytology of head and neck masses. Otolaryngology. 1978;
86:376-381. 4. Sismanis A, Merriam J, Yamaguchi KT, Shapshay SM, Strong MS. Diagnostic value of fine needle aspiration biopsy in neoplasms of the head and neck. Otolaryngol Head Neck Surg. 1981; 89:62-66. 5. Feldman PS, Kaplan MJ, Johns ME, Cantrell RW. Fine needle aspiration in squamous cell carcinoma of the head and neck. Arch Otolaryngol.
1983;109:735-742. 6. Eichhorn T, Schroeder HG, Glanz H, Scherk WB. Histologically controlled comparison of palpation and sonography in the diagnosis of cervical lymph node metastasis. Laryngorhinootologie.
1987;66:266-274.
7. Baatenburg de Jong RJ, Rongen RJ, de Jong PC, Lam\l=e'\risJS, Knegt P. Screening for lymph nodes in the neck with ultrasound. Clin Otolaryngol. 1988;13:5-9. 8. Baatenburg de Jong RJ, Rongen RJ, Lam\l=e'\ris JS, Harthoorn M, Verwoerd CDA, Knegt P. Metastatic neck disease: palpation vs ultrasound examination. Arch Otolaryngol Head Neck Surg. 1989;115:689-690. 9. Som PM. Lymph nodes of the neck. Radiology. 1987;165:593-600. 10. Tubman DE. Newer techniques for the radiographic diagnosis of head and neck malignancy. In: McQuarrie DG, ed. Head and Neck Cancer. St Louis, Mo: Mosby-Year Book; 1986:37\x=req-\
53. 11. Martin H. Untimely lymph node biopsy. Am J Surg. 1962;102:17-18. 12. McGuirt WR, McCabe BF. Significance of node biopsy before definitive treatment of cervical carcinoma. Laryngoscope. 1978;88:594-599. 13. Engzell U, Esposti PL, Rubio C, Sigurdson A, Zajicek J. Investigation on tumor spread in connection with aspiration biopsy. Acta Radiol Ther Phys Biol. 1971;10:385-399. 14. Lindberg LG, Akerman M. Aspiration cytology of salivary gland tumors: diagnostic experience from 6 years of routine laboratory work.
Laryngoscope. 1976;86:584-594. 15. Feldman PS, Kaplan MJ, Johns ME, Cantrell RW. Fine needle aspiration in squamous
cell carcinoma of the head and neck. Arch Oto-
laryngol. 1983;109:735-742. 16. Zajicek J. Aspiration Cytology, I: Cytology of Supradiafragmatic Organs. New York, NY: S Karger; 1974:67-89, 97-107. Monographs in Clinical Cytology. 17. Peters BR, Schnadig VJ, Quinn FB, et al. Interobserver variability of fine-needle aspiration of head and neck masses. Arch Otolaryngol Head Neck Surg. 1989;115:1438-1442.
Downloaded From: http://archotol.jamanetwork.com/ by a University of Pennsylvania User on 06/15/2015