PART TWO
Diagnostic Cytology
in adult patients.17 Clinical thyroid cancer has a prevalence of
approximately 2.5 cases per 1000 persons,18 but occult thyroid
cancer is found at autopsy with a prevalence of 36 cases per
1000,19 indicating that only 1 of 15 thyroid cancers is clinically
manifested. Among surgically excised cold nodules, the fre-
quency of malignancy has been reported to range from 1.5 to
38%.20 This wide variability depends largely on the criteria used
in selecting patients for surgery, but other factors such as sex, age,
genetics, race, and environment also have an important role.19,21
When the incidence of malignant tumors in surgical samples
excised according to classic criteria (clinical history, palpation,
sonography, and scintigraphy) is compared with that of tumors
excised according to cytologic indications, there is a net gain in
the number of tumors diagnosed in cytologically indicated exci-
sions and an appreciable reduction in the economic cost of the
disease.22 With FNA, the number of thyroidectomies has been
halved, whereas the incidence of malignant lesions has dou-
bled;9,23 therefore, the incidence of differentiated thyroid cancer
has doubled in the past 15 years, being now among the most
rapidly increasing types of cancer. The consequences of this
conservative methodology have yet to be evaluated definitively
because certain benign diseases (nodular goiter, lymphocytic
thyroiditis) coexist with, or eventually develop into, follicular
carcinoma, papillary carcinoma, or lymphoma.24
Methods for assessing thyroid nodules include:
• Clinical history and exploration;
• TSH levels;
• FNA;
• Ultrasound (US); and
• Thyroid scan.
Most patients referred for thyroid FNA present slowly grow-
ing thyroid enlargement and are women; in a few cases, a sud-
den increase size or a painful gland is the main complaint. On
palpation, firm nodules are usually suspicious for carcinoma,
whereas fluctuant or soft nodules suggest a benign process.
Although high-quality FNA has been demonstrated to be more
diagnostically useful and cost effective than any other form of
investigation,25,26 it still is underused in common clinical prac-
tice.27 The most accepted opinion has been that all palpable soli-
tary or dominant nodules in a multinodular goiter should be
subjected to FNA as the initial procedure because thyroid func-
tion tests are usually normal in tumors.15 Despite this prevalent
opinion, current management guidelines for patients with thy-
roid nodules call for use of FNA integrated with other explora-
tory methods, and it is recommended after TSH evaluation and
only in patients with normal or high serum levels.28-30 In the ini-
tial diagnostic evaluation, radionuclide scans, CT scan or MRI,
and testing for antithyroid antibodies or T4, free T4, and T3 are
not indicated.
High-resolution US is the most sensitive method for detect-
ing thyroid lesions, measuring their dimensions, and identifying
their structure. During the past two decades, the widespread use
of US has demonstrated that approximately 90% of all thyroid
nodules are not palpable and that their prevalence increases with
age. In patients with a palpable nodule, additional nodules were
found in 20-40% of cases. Moreover, it has been shown that
neither the number of thyroid nodules nor their sizes are predic-
tive criteria for malignancy;13,31,32 the incidence of thyroid cancer
is similar in patients with a clinically apparent solitary nodule
or multiple nodules. US should not be performed on an other-
wise normal thyroid gland nor used as a substitute for a physical
examination; it is, however, highly recommended as the ini-
tial procedure in high-risk patients (history of familial thyroid
cancer or childhood cervical irradiation) even if palpation yields
normal findings and for those with adenopathies suggestive of a
malignant lesion. In patients with palpable nodules (solitary or
multinodular goiter), it also should be performed—especially
in those with normal (0.5-5.0 mU/L) or high (>5.0 mU/L)
TSH levels—to look for coincidental nodules, detect US features
suggestive of malignancy, and obtain an objective measure of
the baseline volume and characteristics of lesions that will be
assigned to follow-up or medical therapy.28,30
A radionuclide thyroid scan should be recommended when
TSH levels are subnormal (<0.5 mU/L) to document whether
the nodules are functioning (hot), isofunctioning (warm), or
nonfunctioning (cold). If the palpable nodule is hot, FNA is not
necessary.28,30
FNA Technique
The technique of FNA of the thyroid follows general FNA guide-
lines in other organs. Preferably, 21- to 25-gauge needles are used
with disposable 10- or 20-cc syringes mounted on a mechanical
device to facilitate aspiration. The use of smaller gauge needles
does not increase diagnostic accuracy and can make reading the
slides difficult because of greater hemorrhagic contamination.33
To obtain an optimal yield, certain recommendations should
be taken into account. The patient should be placed in a supine
position with the neck extended; a pillow under the patient's
neck is helpful. Local dermal anesthesia is optional but may be
recommended if several aspirations are to be performed. The
aspiration should be carried out by a physician experienced in
palpating thyroid nodules, who should stand on the side oppo-
site the site of the lesion. For better identification of the thyroid
mass, palpation should be performed as the patient swallows.
After selection of the area for the aspiration, the overlaying skin
is sterilized by an alcohol pad. During the puncture, the patient
should remain immobile, holding the breath. The gland is immo-
bilized against the trachea with one hand as the aspiration is
done rapidly and gently, moving back and forth several times for
short distances in one channel and maintaining a negative pres-
sure (5 to 10 mL suction) until blood appears at the cone of the
needle. A common misconception is that the needle should be
moved in different directions through the target; this approach
is often painful and results in bloody aspirates.34 The number
of aspirations made in each patient depends on the nature of
the lesion and the experience of the operator. Generally speak-
ing, one to four aspirations suffice in single nodular lesions
measuring less than 3 cm in diameter. When the lesion is larger,
four to eight aspirations may reduce considerably the number
of false-negative results.26 Nonetheless, it should be noted that
more than one aspiration per nodule does not necessarily yield
more diagnostic material and can cause more trauma and hem-
orrhage. The central zones should be avoided because they often
have regressive changes. In multinodular goiter, several nodules
should be sampled. When the lesion is cystic, an attempt should
be made to extract its contents, followed by palpation of the
thyroid to search for intracapsular nodular lesions.
In certain centers, fine-needle sampling without aspiration
(cytopuncture) has been reported to yield results as good as
or better than conventional FNA.35-37 This technique, based on
the principle of capillarity, seems to be particularly useful in
634
previous page 626 ComprehensiveCytopathology 1104p 2008 read online next page 628 ComprehensiveCytopathology 1104p 2008 read online Home Toggle text on/off