Michael M. Kaplan
Many people have thyroid nodules. Estimates of the prevalence depend on the method of detection. Combined data from several studies in which ultrasonography was used reveal a roughly linear increase in the prevalence of thyroid nodules from near zero at age 15 years, to 50% by about age 60 to 65 years, and to even higher in older people (1). Only about 10% of these nodules are palpable, even by experienced examiners (1,2). The dilemma for the physician is to identify the few patients who have thyroid nodules that need treatment from the vast majority in whom the nodules pose no harm. Optimally, the selection process will have high diagnostic accuracy, minimize medical costs and patient anxiety, and lead to treatment that will improve the patient's health.
The most important consideration in evaluating patients with thyroid nodules is whether the nodule is a thyroid carcinoma. In addition, some thyroid nodules, although benign, reflect processes that can cause thyrotoxicosis or, rarely, hypothyroidism, or require treatment because their size causes an unacceptable appearance or compression of nearby structures in the neck and upper chest. This chapter discusses solitary thyroid nodules, the most important types of which are listed in Table 73.1. However, the distinction between solitary thyroid nodules and multiple nodules is artificial, because about 50% of patients thought on the basis of physical examination to have a solitary nodule have multiple nodules when studied by ultrasonography (2,3). Moreover, the likelihood that an individual nodule is a carcinoma is not much less in patients with a multinodular goiter than it is in patients with a solitary nodule. With respect to thyroid carcinoma, the concern is primarily for differentiated tumors of follicular cell origin, because they are far more common than medullary or anaplastic carcinoma (see Chapters 71 and 72). Patients with the latter types of carcinoma (and thyroid lymphoma) may present in the same way, and fine-needle aspiration biopsy may provide the diagnosis in the same way as in patients with differentiated thyroid carcinoma.
TABLE 73.1. TYPES OF THYROID NODULES
Adenomatous nodule (also called colloid nodule and hyperplastic nodule)
Thyroglossal duct cyst
Carcinomas of thyroid follicular-cell origin
Carcinoma of C-cell origin
Cancer metastatic to the thyroid
Parathyroid cyst and adenoma
Branchial cleft cyst and other epithelial cysts
Most patients with thyroid carcinoma present with an asymptomatic thyroid nodule, as do most patients with benign thyroid nodules. Some clinical findings (Table 73.2) have long been thought to increase the likelihood that a nodule is a carcinoma (1,4). However, the relationship between most of these findings and the presence of carcinoma is not strong. Also, most studies of these factors predate the widespread use of thyroid ultrasonography and biopsy. Some more recent studies are reviewed below. Lastly, there are few data relating clinical variables and cancer risk in patients with small thyroid nodules discovered incidentally.
TABLE 73.2. CLINICAL FINDINGS RELATED TO RISK OF CARCINOMA IN A THYROID NODULE
Age < 20 or >60 years
Men > women (in terms of proportion of nodules that are a carcinoma)
History of benign or malignant thyroid disease
History of other cancer, especially kidney, breast, and colon
History of head or neck irradiation
Symptoms of possible compression of structures adjacent to the thyroid gland:
Hoarseness or other change in voice
Family history of benign or malignant thyroid disease, pheochromocytoma, hyperparathyroidism, familial adenomatous polyposis coli, Gardner's syndrome, Cowden's disease
Large size of nodule
Firm or hard nodule
Pain or tenderness in or around the nodule
Growth during thyroid hormone treatment
A thyroid nodule is more likely to be a thyroid carcinoma in patients under age 20 years and those over 60 years of age than in those in between, but overall most patients with thyroid carcinoma are in the age group of 20 to 60 years. Benign thyroid nodules are four to five times more common in women than men, but thyroid carcinomas are only two to three times more common in women; thus, a nodule is more likely to be a carcinoma in a man. A history of a benign thyroid nodule or goiter may increase the risk that a new nodule is a carcinoma, but ascertainment bias probably explains this increase. Hashimoto's thyroiditis can cause nodules, as a result of scarring and localized areas of hyperplasia or lymphocytic infiltration, and in addition it strongly predisposes to the rare thyroid lymphomas (see Chapter 72). Cancer metastatic to the thyroid gland is rare; when present, the primary tumor is usually a renal cell, colon, or breast carcinoma.
Radiation therapy directed to the head, neck, and upper chest increases the risk of both benign thyroid nodules and thyroid carcinoma (see section on pathogenesis in Chapter 70). In the past, most patients with radiation-related thyroid carcinoma had received radiation therapy for benign conditions, such as thymic or tonsillar enlargement, acne, or birthmarks in infancy, childhood, or adolescence. Radiation therapy for these conditions largely ceased 30 to 40 years ago; now radiation-related thyroid carcinoma occurs mostly in patients who received radiation therapy for lymphoma or head and neck cancer, although a high incidence of thyroid carcinoma occurred in children exposed to fallout from the Chernobyl nuclear power plant explosion in 1986. Although some studies suggest that up to 30% of patients with thyroid nodules and a history of radiation exposure in early childhood who undergo surgery have thyroid carcinoma, some of the carcinomas were incidental microcarcinomas distant from the nodule that prompted the surgery.
Large thyroid nodules, benign or malignant, can distort surrounding structures in the neck or upper mediastinum, causing difficulty breathing, shortness of breath or cough from tracheal compression, or difficulty swallowing from esophageal compression. The presence of these symptoms may increase the likelihood that a nodule is a carcinoma, but most carcinomas are small and do not cause these symptoms. Similarly, hoarseness or a change in voice can be caused by infiltration or compression of the recurrent laryngeal nerve by a thyroid carcinoma, but other causes of a change in voice are much more common.
The family history is sometimes relevant. From 4% to 8% of patients with papillary thyroid carcinoma have a family history of the same tumor, although the genetics are not understood. In familial adenomatous polyposis and its variant, Gardner's syndrome, and in Cowden's disease, also called the multiple hamartoma syndrome, the prevalence of benign thyroid nodules and thyroid carcinoma is increased (see section Pathogenesis in Chapter 70). Medullary thyroid carcinoma can be familial (see Chapter 71), as part of the multiple endocrine neoplasia type 2 and related syndromes, in which there is a very high penetrance of medullary carcinoma, and less often pheochromocytoma and primary hyperparathyroidism.
Large nodules, nodules fixed to the surrounding tissue, and very firm nodules tend to be carcinomas, but some benign nodules have a calcified shell that makes them very hard. Some thyroid carcinomas are tender or cause pain, but these findings are much more likely to be caused by hemorrhage into a benign nodule. Subacute granulomatous thyroiditis is another cause of thyroid pain, usually bilateral but occasionally starting in one lobe and then affecting the other lobe. Rapid growth of a nodule is of concern, but the sudden appearance or increase in size of a nodule in a day or less is almost always due to hemorrhage into a benign nodule. Growth of a nodule during follow-up or thyroid hormone treatment has traditionally been considered to suggest that the nodule is a carcinoma, but benign nodules also grow (5). Conversely, some carcinomas do not grow during follow-up. The presence of cervical lymphadenopathy suggests that a thyroid nodule is a carcinoma.
Some clinical risk factors for thyroid carcinoma have been analyzed in several studies of patients who had surgery, partly on the basis of suspicious needle biopsy results (6,7,8,9,10,11). Four of these studies were retrospective, and none evaluated the putative risk factors prospectively in a separate set of patients, either in patients with inconclusive or suspicious biopsy results or those in whom biopsy revealed only benign thyroid cells. Despite these limitations, the studies consistently indicate that nodules that are fixed to surrounding structures and nodules that are larger than 4 cm in diameter are likely to be carcinomas. In three studies, age less than 20 years was a risk factor for carcinoma. The results were inconsistent with respect to whether the risk for carcinoma was higher in men, patients over 60 years of age, patients with rapidly growing nodules, those with solitary or multiple nodules, and patients in whom the nodule was firm in consistency.
Serum Hormone and Other Measurements
Serum thyrotropin (TSH) should be measured in all patients with thyroid nodules (Table 73.3), to determine if the patient has an autonomously functioning thyroid adenoma causing subclinical thyrotoxicosis, and to identify unsuspected hypothyroidism, which may or may not be related to the patient's nodules. Measurements of serum antithyroid peroxidase antibodies may be indicated to distinguish between Hashimoto's thyroiditis and Hürthle-cell tumors; biopsies of both may contain many Hürthle cells. Measurements of serum thyroglobulin have little value in patients with thyroid nodules; the values may be normal or high in patients with either benign nodules or carcinoma.
TABLE 73.3. DIAGNOSTIC TESTS FOR THYROID NODULES
TSH, for assessment of thyroid function
Antithyroid antibodies, for identification of autoimmune thyroiditis
Calcitonin, for identification of medullary carcinoma
Magnetic resonance imaging
Positron emission tomography
Measurement of serum calcitonin is indicated in patients with a nodule if there is a family history of medullary thyroid cancer or other components of the multiple endocrine neoplasia type 2 syndrome, or a family history of thyroid carcinoma in which the type is not known. Some have recommended that serum calcitonin be measured in all patients with thyroid nodules, to identify patients with sporadic medullary carcinoma. In a study of 10,864 patients with thyroid nodules, 0.4% had high serum calcitonin concentrations and proved to have medullary thyroid carcinoma (12). However, about two thirds of those patients had suspicious needle biopsy results, and only 15 medullary thyroid carcinomas were identified solely by measurement of serum calcitonin (1 in 724 patients). In other, similar studies, measurements of serum calcitonin have identified patients with medullary carcinoma, but other patients with high serum calcitonin concentrations did not have medullary carcinoma (13). Because of the possibility of false-positive results and the lack of standardization of the measurement, it has not been widely adopted in the United States.
About 5% to 10% of thyroid nodules are autonomously functioning thyroid nodules, nearly all of which are adenomas (14). In some of these tumors, activating mutations of the TSH receptor (15) or the G (guanine nucleotide-binding) protein that links the TSH receptor with adenylyl cyclase (16) cause the TSH-independent function. In patients with thyroid nodules who have low serum TSH concentrations, a thyroid scintiscan, preferably with iodine 123, should be done to determine if the nodule is hyperfunctioning, that is, if it concentrates the radionuclide more than the rest of the thyroid gland (a so-called “hot” nodule) (see Chapter 12). If so, biopsy is not necessary unless there are findings suspicious of carcinoma, such as rapid growth or hypofunctioning areas within the nodule. In the past, a scintiscan was usually the first imaging study done in patients with thyroid nodules. However, since hyperfunctioning nodules are uncommon and scintiscans do not distinguish between benign and malignant hypofunctioning nodules, a scintiscan should be performed before needle biopsy only in patients with low serum TSH concentrations. If the practice is to biopsy all nodules, a scintiscan may be indicated in patients in whom the biopsy reveals a follicular tumor, because some autonomously functioning adenomas have this cytology (17), do not secrete enough thyroid hormone to lower TSH secretion, especially if less than 2.5 cm in diameter (14), and yet are active enough to be revealed as hyperfunctioning by scintiscan. The risk for carcinoma is low in this subset of patients, just as it is in those patients with an adenoma who have low serum TSH concentrations.
Thyroid ultrasonography is now widely used for evaluation of patients with thyroid nodules (see Chapter 16), although the most appropriate role of this test in the evaluation of patients with an apparently solitary thyroid nodule is not yet certain. The use of ultrasound guidance to perform needle biopsies of thyroid nodules is discussed later. Those who advocate diagnostic ultrasonography in all patients with a solitary nodule on physical examination argue that about 50% of the patients have other, nonpalpable nodules (4), and that some or even many of them are large enough to warrant biopsy, the latter depending on the size threshold considered to warrant biopsy.
For the most part, the ultrasonographic features of benign nodules and carcinomas are similar, but certain features increase the likelihood that a nodule is a carcinoma (18). These features are diffuse microcalcifications, hypoechogenicity, an irregular margin indicative of invasive growth, an irregular sonolucent halo or no halo around the nodule, and regional lymphadenopathy. However, calcifications and invasiveness are uncommon, even in malignant nodules. A search for lymph nodes is often omitted, but should be done routinely. Findings that make it unlikely that a nodule is a carcinoma are the presence of a well-defined sonolucent halo and a regular margin, and coarse calcification within the nodule or a rim of calcification surrounding the nodule. However, there are no ultrasound characteristics that definitively exclude carcinoma and therefore render biopsy unnecessary. Nonetheless, the ultrasonographic characteristics of a nodule may be of use in estimating cancer risk in patients in whom needle biopsy was inadequate and those who are hesitant to undergo or prefer to delay biopsy.
Thyroid ultrasonography has important limitations (18,19). The procedure is very operator dependent, and ultrasound reports often lack crucial information, such as the dimensions of nodules in all three planes (length, width, and depth), the location of a nodule in the thyroid lobe, comments about density or calcification, the appearance of the border of the nodule, and the presence of lymph nodes. Ultrasonography is usually performed as a diagnostic test, with no intent on the part of the ordering physician to have a biopsy done or no capability for biopsy at the ultrasound facility. Patients are better served if ultrasonography is done under conditions in which biopsy can be performed immediately if indicated. The operator must be interested in and knowledgeable about thyroid nodules, know what to look for, and be able to use the results to perform needle biopsy at the same time.
Computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography (PET) using 19-fluorodeoxyglucose have no established role in the evaluation of patients with thyroid nodules. These tests are listed in Table 73.3 only because in some patients thyroid nodules are discovered when the test was performed for other reasons. A single thyroid nodule identified on a PET scan may be a carcinoma, but some benign nodules and Hashimoto's thyroiditis take up substantial quantities of 19-fluorodeoxyglucose as well.
Fine-needle biopsy, which yields single cells or clumps of cells, is the standard test to determine whether a thyroid nodule is benign or malignant, and the test also has other uses (Table 73.4). It is often called fine-needle aspiration biopsy, but the more general term will be used here because aspiration by negative pressure is not always used. The primary goal of fine-needle biopsy is to reduce the number of unnecessary thyroid operations, by identifying those nodules that are benign or at least unlikely to be malignant. For this purpose, fine-needle biopsy is far more accurate than any other available test or combination of tests. Before the procedure was introduced, about half of patients with palpable thyroid nodules were referred for operation, but only about one fourth of those who underwent surgery had thyroid carcinoma. Hence, many patients with benign nodules were subjected to the costs, morbidity, and, rarely, risk for mortality of surgery (20,21).
TABLE 73.4. USES OF THYROID NEEDLE BIOPSY
Selection of therapy for a thyroid nodule
Surgery versus observation
Extent of surgery
Diagnosis of the cause of thyroid enlargement
Removal of fluid or blood from a cystic nodule
Injection of a sclerosing agent for a recurrent cyst
Ablation of a nodule by ethanol injection or laser coagulation
Postoperative evaluation of cervical lymph nodes
Diagnosis of infective thyroiditis
The use of fine-needle biopsy can reduce the proportion of nodules for which surgery is advised to less than 20%; at the same time, the proportion of resected nodules that are malignant is 50% or even higher. When an adequate sample is obtained, the reliability of fine-needle biopsy for excluding carcinoma is 98% to 99%. The second main use of fine-needle biopsy is to help plan the extent of surgery. Large-needle biopsy, which provides cores or fragments of tissue for conventional histologic evaluation, is rarely used now. Details of the methods can be found elsewhere (22).
Selection of Patients
Fine-needle biopsy is the most appropriate test for almost all patients with thyroid nodules, the exception being patients suspected to have an autonomously functioning thyroid adenoma. Biopsy is usually advised if the maximal diameter of the nodule is 1 cm or higher (23,24), but some researchers suggest a threshold of 1.5 or 2.0 cm (25,26). This distinction depends in part on the method of ascertainment. Nodules smaller than 1.5 cm are often not palpable, whereas much smaller nodules can be detected by ultrasonography. Even if ultrasonography is performed, or the nodule is detected incidentally when an imaging study of the neck is done for another purpose, nodules smaller than 1 cm in diameter are often not biopsied. The rationales for this threshold are that smaller nodules are so common that it is impractical to refer all of them for biopsy, and the risk for missing a carcinoma that will have adverse health consequences is low. Many people have small thyroid carcinomas that never become evident during life and cause no health problems. The prevalence of thyroid carcinoma at autopsy is about 4% in the United States (25), where there are currently about 2.5 million deaths per year, fewer than 24,000 incident cases of thyroid carcinoma diagnosed per year, and fewer than 1,500 deaths from thyroid carcinoma per year (27). Furthermore, among patients with thyroid carcinoma, the outcome is strongly influenced by the size of the tumor at the time of diagnosis: the life expectancy of patients with a tumor that is smaller than 1 cm in diameter is the same as that of the general population (see section Radioiodine and Other Treatments and Outcomes in Chapter 70).
As noted above, incidental thyroid nodules may be detected by ultrasonography or other imaging procedures performed to evaluate neck structures other than the thyroid gland, or by thyroid ultrasonography done to evaluate neck pain, difficulty swallowing, or a known thyroid abnormality. Most of these nodules are too small to be detected by palpation, Others are located in the middle of a thyroid lobe (and do not distort its shape), or behind or below the lobe (28); have a soft consistency; or are hidden by subcutaneous fat, prominent neck muscles, or kyphosis of the cervical spine. These factors are independent of the biology of the nodule, and therefore it is likely that a nonpalpable nodule is a carcinoma as often as is a palpable nodule of the same size. The available data support this conclusion. Among patients with nonpalpable, incidentally discovered thyroid nodules who have undergone fine-needle biopsy—by necessity ultrasound-guided—4% to 10% have had thyroid carcinoma (3,29,30,31,32), very similar to the results in patients with palpable nodules. Therefore, the evaluation, including the size cut-off for biopsy, should be the same no matter how the nodule was detected.
Fine-needle biopsy is advisable in patients with a multinodular goiter who have a discrete hypofunctioning nodule within the goiter or a nodule of uncertain functional status that is growing, and patients with a partially cystic nodule after fluid has been removed. The prevalence of carcinoma in these nodules is similar to that in solitary solid nodules (3,6,7). Fine-needle biopsy is also indicated when a hypofunctioning area is seen in an otherwise hyperfunctioning nodule, although the great majority of these areas represent central degeneration within the nodule, not carcinoma. A few patients with thyrotoxicosis caused by Graves' disease also have hypofunctioning nodules, whether detected by palpation or scintiscan; these nodules should be biopsied because some are carcinomas. Patients treated with head and neck radiation for nonthyroid conditions, or accidentally exposed to radiation, who have thyroid nodules should be evaluated by biopsy like any other patient; biopsy results are as accurate in these patients as in patients without such a history (33,34).
Pregnant women found to have a thyroid nodule should undergo biopsy, like any other patient, for two reasons. First, the detection of the nodule raises concern about carcinoma that can usually be alleviated simply and quickly. Second, biopsy facilitates decisions regarding surgery in women whose nodules should be excised, including the options of surgery in the second trimester, as soon as possible after delivery, or at a time most consistent with plans for breast-feeding (35,36).
Fine-needle biopsy may not be as accurate in children as in adults, as discussed later.
Technique of Fine-Needle Biopsy
About two thirds of thyroid specialists in the United States currently perform fine-needle biopsy guided by palpation, whereas about two thirds in Europe use ultrasound guidance (37). The routine use of ultrasonography increases the expense of the procedure, usually requires advance scheduling, and requires an assistant. In a study from a single institution comparing the diagnostic accuracy of palpation-guided fine-needle biopsy of nearly 5,000 thyroid nodules with ultrasound-guided fine-needle biopsy of a similar number of nodules (38), the diagnostic accuracy was better when ultrasonography was used, but the differences between the two groups were small, and patients were not randomly allocated to each group. The biggest difference was in the frequency of biopsies in which the sample was inadequate; it was 3.5% for the ultrasound-guided biopsies versus 8.7% for the palpation-guided biopsies. The results were falsely negative (missed carcinoma) in 1% of nodules examined via biopsy with ultrasound guidance and in 2.3% of the nodules examined via biopsy with palpation guidance, although there were more nodules smaller than 1 cm in the ultrasound-guided biopsy group.
Palpation-guided fine-needle biopsy is satisfactory for nodules that are easily palpable. For nodules difficult to palpate, and of course those that are not palpable, the biopsy must be guided by ultrasonography. If palpation-guided biopsy yields an inadequate sample, an ultrasound-guided biopsy should be performed, especially if the nodule is relatively small. The alternate approach of doing only ultrasound-guided biopsies is favored by some physicians.
The actual procedure is performed with the patient supine, with a pillow under the shoulders to facilitate neck extension. The patient should be asked not to talk or swallow while the needle is in the neck. The skin is cleaned with alcohol, and the skin where the biopsy needle is to be inserted is infiltrated with 1 to 2 mL of 1% lidocaine (this may be omitted). The nodule is fixed by the fingers of one hand, and the needle in the other hand is inserted perpendicular to the anterior surface of the neck. Twenty-five-gauge, 1½-inch needles provide excellent specimens and are less likely than larger needles to cause bleeding that dilutes the specimens. The ease with which bleeding is induced depends on whether the nodule is dense and relatively avascular, or loosely organized and more vascular. Occasionally, 22- or 23-gauge needles may be needed to obtain adequate specimens from hard, fibrotic nodules.
The biopsy sample may be obtained without or with suction. In the nonsuction technique, the needle is gripped like a pencil, which facilitates precise needle placement for small nodules, and allows the needle to be moved both in and out over a few millimeters and rotated (39). This combined motion uses the bevel of the needle for cutting, which frees cells that flow into the needle by capillary action while the needle is held steady for about 10 seconds. Material can usually be seen entering the hub of the needle. Samples obtained in this way are less likely to be diluted with blood than samples obtained using suction. If the nonsuction method does not yield a satisfactory sample, or if cyst fluid requires removal, suction is used.
For the suction technique, a 10-mL syringe is used. Some operators put the syringe in a pistol grip designed to facilitate a one-handed biopsy. After the nodule is punctured, the plunger is withdrawn about two thirds of the way, and the needle is moved repeatedly in and out over a distance of 2 to 4 mm. At the first appearance of fluid, negative pressure is released and the needle is withdrawn. No fluid should enter the syringe. If this happens, the specimen may be too dilute and partly lost in the syringe. Material may appear in the syringe if too large a needle is used, if negative pressure is too vigorous, if the nodule is extensively degenerated or unusually vascular, or if the nodule is cystic. For the first two situations, adjustments can be made on subsequent aspirations. For the latter two situations it may help to insert the needle at the periphery of the nodule. The initial aspiration may yield no specimen if the needle is not in the nodule, if the needle is too fine, if negative pressure is not vigorous enough, or if the nodule is fibrotic. If the nodule is purely cystic, it will collapse with aspiration. The fluid should be sent for cytologic examination, although clear fluid is usually acellular. If there is a solid component, any residual mass should be examined via biopsy.
After needle withdrawal, 3 to 4 mL of air is forced through it to expel the specimen onto a glass slide. A grossly satisfactory specimen consists of a small amount of red-orange fluid. Voluminous bloody specimens or specimens that consist of watery, greenish brown fluid, crystal clear jellylike fluid, thick oily fluid, or cheesy white material usually contain insufficient epithelial cells for diagnosis. In some clinics, specimens are stained and examined immediately, so that if the biopsy specimen is inadequate the procedure can be repeated immediately.
Once the specimen is on the slide, it is smeared and fixed immediately to prevent air drying. Some pathologists use air-dried smears stained with May-Giemsa-Grunwald stain, which enhances cytoplasmic detail; some prefer the crisp nuclear detail obtained with the Papanicolaou stain, which requires immediate fixation before air drying occurs; and some use both stains or others. Alternatively, the biopsy specimens can be expelled from the needle directly into a fixative/preservative solution, and slides are later prepared in the laboratory. This permits the use of procedures that concentrate the thyroid cells and eliminate red blood cells. The choice of slide preparation technique needs to be made cooperatively between the person doing the biopsy and the cytopathologist.
Complications of thyroid fine-needle biopsy are exceedingly rare. Biopsy with a 25-gauge needle is safe even in patients taking anticoagulant drugs. Nevertheless, patients should be advised that trauma to adjacent structures is possible. Occasionally, a small hematoma forms at the biopsy site. An ice pack and pressure are adequate treatment. Occasional patients have mild pain that may radiate to the jaw or ear for a day or two, or rarely longer. Infection is rare. Seeding of carcinoma cells in the needle track has been reported in only three patients; two of the patients had large-needle biopsies (34) and one patient had a fine-needle biopsy (40). The reported seedings were excised and had no unfavorable impact on prognosis. Serum thyroglobulin concentrations may increase for a few days after a biopsy (41).
For adequate cytologic evaluation, the biopsy should yield at least six clusters of 20 or more thyroid cells on each of at least two separate aspirates. The guidelines for thyroid fine-needle biopsy of the Papanicolaou Society of Cytopathology cite this as one criterion of an adequate biopsy; alternate criteria are five to six clusters of at least 10 thyroid cells, or 10 clusters of at least 20 thyroid cells (42). When biopsy specimens can be examined immediately, sampling can stop as soon as adequate cellular material is collected. If immediate examination is not possible, sufficient aspirations to provide six separate specimens that appear grossly satisfactory yield adequate results in a high percentage of patients (43), although in many centers only two to four aspirations are performed routinely. For successive samples, the needle is aimed at a different part of the nodule, if size permits. A single aspiration with just one cluster of obviously malignant cells may be sufficient to diagnose a carcinoma, but many more cells are needed to exclude carcinoma, because benign thyroid cells adjacent to a carcinoma may be obtained or the needle may be misplaced. Also, within a carcinoma, the cells in some areas may look more normal than the cells in other areas.
Despite multiple aspirations, specimen cellularity is still sometimes unsatisfactory, and the biopsy is considered inadequate. In such patients, the biopsy should be repeated. A second attempt guided by palpation of a readily palpable nodule is successful in about half the patients, and ultrasound-guided biopsy was successful in 63% of second biopsies in one series (44). If two attempts fail, a third is occasionally successful (44).
After the biopsy, the patient's skin is dressed with an adhesive bandage, and the patient may depart after a few minutes of observation to let lightheadedness pass and to check for bleeding or local swelling.
Results of Fine-Needle Biopsy
Examples of the most common cytologic findings from fine-needle biopsies of thyroid nodules are shown in Fig. 73.1. The results of fine-needle biopsies are reported in various ways, and it is important that the clinician understand the terminology used by the cytopathologist. Inadequate biopsies, as noted above, are those in which too few cells are obtained to allow any interpretation. Other terms used include or or , and or (see Chapter 21). Some cytopathologists use the terms or instead of or , because follicular adenomas and well-differentiated follicular carcinomas cannot be distinguished reliably by any cytologic criteria. should not be confused with normalbenign, papillary carcinomasuspicious for papillary carcinomafollicular neoplasmfollicular tumornondiagnosticinconclusivefollicular neoplasmtumorNondiagnosticinadequate.
FIGURE 73.1. Photomicrographs of fine-needle biopsies of the most common types of thyroid nodules. A. Colloid nodule, colloid-rich type (magnification x400). B. Degenerating colloid nodule, colloid-poor type (x40). C. Hashimoto's thyroiditis (x400). D. Follicular tumor (x400). Most follicular tumors are adenomas, but the possibility of follicular thyroid carcinoma cannot be ruled out. E. Hürthle-cell nodule showing cells with abundant granular cytoplasm (x400). Some of these nodules are adenomas, others are carcinomas, and still others are aggregations of Hürthle cells in patients with Hashimoto's thyroiditis. F. Papillary carcinoma (x40). These photomicrographs were kindly provided by Edward Bernacki, M.D.
Table 73.5 shows the distribution of cytologic diagnoses in large studies of palpation-guided biopsies, studies of at least 100 patients with one or more palpable and nonpalpable nodules in whom all biopsies were ultrasound guided, and studies of nodules that could be examined via biopsied only with ultrasound guidance because they were difficult or impossible to palpate (3,30,31,32,37,38,45,46,47,48,49,50,51). For each study, the average percentage of nodules in the four cytologic categories is similar; overall, 74% of the biopsies revealed benign cells, 4% revealed malignant cells, 10% were suspicious (follicular lesions), and 11% were inadequate. The variability of results between series is also reasonably similar within each of the three groups, notwithstanding differences in patient populations, criteria used for an adequate biopsy, and criteria used to differentiate between benign thyroid cells and follicular tumors. Even with repeated fine-needle biopsy and routine use of ultrasonography, there is probably an irreducible minimum of about 3% to 5% of nodules from which adequate fine-needle biopsy samples cannot be obtained.
TABLE 73.5. DISTRIBUTION OF CYTOLOGIC RESULTS IN REPRESENTATIVE THYROID FNB SERIES
Number of nodules
Inconclusive or Suspicious %
Unsatisfactory (Inadequate) %
Guided by palpation
adapted from ref 45*
Guided by ultrasound, palpable and non-palpable nodules
Guided by ultrasound, difficult to palpate or non-palpable nodules
*Pooled data from several centers excluding the Mayo Clinic, from which more recent data (47) are listed separately. The values in some rows do not add to 100% because of rounding.
Standard calculations of sensitivity, specificity, and false-positive and false-negative rates are complicated, and probably not of great utility. The reasons are that the great majority of patients with nodules with benign cytology do not undergo surgery; therefore, there is no reference standard for a benign final diagnosis, and on average about 10% to 12% of biopsies yield inadequate samples. In practical terms, about 2% to 3% of benign nodules, as determined by fine-needle biopsy, subsequently prove to be carcinomas. Conversely, about 2% to 5% of malignant nodules, as determined by fine-needle biopsy, prove to be benign on histologic study. Because a fine-needle biopsy diagnosis of follicular tumor does not exclude carcinoma, this biopsy result should only be considered as a guide to identify patients for whom surgery is probably advisable (52). There are as yet no cellular or molecular markers proven to distinguish reliably between follicular adenomas and follicular carcinomas.
There have been two studies of fine-needle biopsy in children with thyroid nodules. In the larger study, of 60 children, 50 of 51 nodules (98%) for which the biopsy revealed benign thyroid cells proved to be benign, and 8 of the 9 nodules (89) that proved to be carcinomas were identified by biopsy (53). In the smaller study, benign needle biopsy diagnoses were correct in only 4 of 7 cases (57), and only 2 of 5 thyroid carcinomas (40) were identified by biopsy (54). Possible reasons that these results are not as good as those in adults include differences in patients, level of experience of the persons doing the biopsies, lack of cooperation of the children leading to fewer samples, and small numbers of patients.
Limitations of Fine-Needle Biopsy of Thyroid Nodules
There are some limitations of fine-needle biopsy of thyroid nodules (Table 73.6). The principal limitation is inexperience, both in obtaining adequate specimens and in interpreting the specimens. Instructional material is available, but there is no substitute for experience. Fine-needle biopsy is not applicable to all nodules. Some are too small and too inaccessible for accurate needle placement, or too far down in the chest to be aspirated safely. Others are so degenerated that useful material cannot be obtained. Finally, a fine-needle biopsy of one nodule provides no information about any other nodules that the patient may have.
TABLE 73.6. LIMITATIONS OF FINE-NEEDLE BIOPSY OF THE THYROID
Inaccurate, vague, or nonspecific reports
High level of interest and experience of clinician and pathologist and good communication between them
Adequate specimens with inconclusive cytology
Consider clinical features and other laboratory and imaging results
Surgery may be appropriate
Repeat the biopsy, with ultrasound guidance
Observation or fine-needle biopsy with ultrasound guidance
Fine-needle biopsy with ultrasound guidance, if practical and safe
False negative and false positive results
Inform patient of possibility
Use stringent criteria for adequacy
Consider clinical features
Perform follow-up evaluations, which may include repeat biopsy
Sampling limitations in large, heterogeneous nodules
Take multiple samples Perform biopsy with ultrasound guidance to sample non-cystic areas and areas of greatest concern according to appearance
Multiple nodules in one thyroid gland
Biopsy all that are accessible and of concern based on clinical and imaging criteria
Extreme patient anxiety about cancer despite biopsy finding of benign nodule
Consider surgery, using biopsy results to help decide on extent of resection
The legal consequences of false-positive or false-negative diagnoses, and therefore in advising patients to undergo, or to not undergo, surgery are of concern to some physicians. Fine-needle biopsy improves diagnostic precision and reduces the potential for error. However, patients should be informed that the procedure is not always successful or reliable.
COMBINED RESULTS OF SERUM, IMAGING, AND FINE-NEEDLE BIOPSY TESTING
For all adults with apparently solitary thyroid nodules, the combined results of serum measurements, imaging studies, and fine-needle biopsies lead to the following estimates:
Malignant nodules constitute approximately 3% to 6% of thyroid nodules at least 1 cm in maximum diameter. Most are differentiated thyroid carcinomas of follicular-cell origin.
Autonomously functioning adenomas account for approximately 5% to 10% of thyroid nodules in the United States. Their prevalence may differ in other geographic areas.
Cystic nodules, with more than 75% fluid content, account for 10% to 15% of thyroid nodules. Simple epithelial cysts are rare. The fluid in most cystic nodules is the result of hemorrhage or thyroid cell degeneration; either is as likely to occur in thyroid carcinomas as in benign nodules. Therefore, this category overlaps with the other types of thyroid nodules.
Benign thyroid nodules that are 75% cystic or less and are isofunctioning or hypofunctioning constitute the remaining 70% to 80%. This category includes adenomatous, colloid, and hyperplastic nodules and follicular adenomas (see Chapter 21).
APPLICATION OF FINE-NEEDLE BIOPSY TO THE MANAGEMENT OF PATIENTS WITH THYROID NODULES
Patients in whom an adequate biopsy reveals benign thyroid follicular cells do not need surgery based on risk for carcinoma. If surgery is chosen because of the appearance of the nodule, or the nodule compresses nearby structures, a lobectomy is usually the appropriate procedure. The chance of serious harm from a false-negative biopsy result in a patient who does not undergo surgery is less than the 2% to 3% frequency of false-negative biopsies, because thyroid carcinomas for which the biopsy was falsely negative are usually low-grade tumors that are still curable even if the diagnosis is delayed. If the patient returns for periodic reevaluation, a repeat biopsy can be performed if the nodule has enlarged. When repeat biopsies are performed, the biopsy diagnosis is more abnormal and suggests the need for surgery in 1% to 3% of patients (5,55,56,57).
A fine-needle biopsy diagnosis of papillary carcinoma is almost always confirmed at operation, and surgery is the initial treatment for patients with this tumor. A fine-needle biopsy diagnosis of medullary carcinoma or undifferentiated or anaplastic carcinoma is also nearly always confirmed at operation, and surgery is advised for patients with potentially resectable lesions. In the case of thyroid lymphoma, also reliably detected by fine-needle biopsy, open biopsy to obtain tissue for precise classification of the tumor using flow cytometry and immunohistochemical techniques is usually indicated.
In patients in whom the fine-needle biopsy diagnosis is suspicious for papillary carcinoma or follicular tumor (including Hürthle-cell tumor), the likelihood of carcinoma is high enough (25%–75) that operation should be advised, unless there is a major contraindication to surgery. With respect to follicular tumors, the guidelines of the Papanicolaou Society of Cytopathology take note of two approaches to reporting cytologic findings: either making no further comment, or subdividing follicular tumors into two or three categories according to the degree of cellular atypia (42). If the cells show little or no atypia, about 5% to 15% of the nodules are follicular carcinomas. This risk is high enough to advise operation for most patients, but an 85% to 95% probability that the nodule is benign might make observation an acceptable choice for older patients or those with medical problems that increase the risk for complications of surgery. Higher degrees of atypia increase the likelihood that the nodule is a follicular carcinoma, and usually lead to a stronger recommendation for surgery. In patients with a fine-needle biopsy diagnosis of follicular tumor, repeating the biopsy rarely clarifies the nature of the nodule (58).
Among patients in whom the fine-needle biopsy was inadequate, thyroid carcinoma was found in an average of 14% of those selected for surgery, or 29 of 210 patients in data pooled from several studies (30,34,38,46,49,59,60). Patients whose biopsy results are inadequate may choose surgery because of fear of carcinoma, or may be advised to have surgery because of the presence of some clinical finding associated with a higher risk for carcinoma (Table 73.2).
In the past, many patients undergoing surgery for a thyroid nodule had a lobectomy (usually including the isthmus) with a frozen section, after which the operation was terminated or extended depending on the results of the histologic findings in the frozen sections of the nodule. This strategy has been largely abandoned; there is usually a firm biopsy diagnosis now, which is more reliable than frozen-section diagnosis (52). Although two recent studies report 78% and 88% accuracy of frozen-section diagnoses of thyroid nodules (61,62), most centers have less favorable results. Frozen-section diagnoses may be difficult to obtain, or the results may be erroneous, and reversed after review of the permanent sections (52). Frozen-section diagnosis is least reliable in distinguishing between follicular adenomas and follicular carcinomas, because this distinction is based primarily on the presence of vascular and capsular invasion by carcinomas.
These findings lead to the following conclusions:
If surgery is chosen despite a fine-needle biopsy diagnosis of a benign nodule, the operation can be a lobectomy; frozen-section analysis is not needed.
If surgery is chosen for a patient with a fine-needle biopsy diagnosis of follicular tumor, an initial lobectomy will be the correct procedure in about 85% of patients. Frozen-section is usually not helpful, as noted above. If the nodule is a small, low-grade follicular carcinoma, lobectomy may be an adequate operation.
For patients with a fine-needle biopsy diagnosis of Hürthle-cell tumor or suspected papillary carcinoma, the nodule is more likely to be a carcinoma, and the patient should be fully informed of the possible need for near-total thyroidectomy before the operation. Frozen-section diagnosis may or may not provide additional guidance. Many patients choose near-total thyroidectomy, accepting the additional risks to avoid the risks and unpleasantness of a second operation.
For patients with a fine-needle biopsy diagnosis of papillary or other carcinoma, the chance of a false-positive biopsy diagnosis is so small that near-total thyroidectomy or, for patients with medullary carcinoma, total thyroidectomy and extensive bilateral lymph node resection is indicated (see Chapter 71). The exact surgical procedure may be guided by frozen-section evaluation of lymph nodes, parathyroid glands, and tumor invasion into fat and muscle.
OTHER USES OF FINE-NEEDLE BIOPSY OF THE THYROID GLAND
Fine-needle biopsy can be used to evaluate diffuse enlargement of the thyroid gland or selective enlargement of one lobe, if serum tests and scintiscanning leave the etiology of the abnormality uncertain, or if the scintiscan shows a discrete area of hypofunction (Table 73.4). The area of concern is surveyed by aspirating six to eight sites, using the same technique as for nodules. Such survey biopsies can help distinguish patients with lymphoma from those with Hashimoto's thyroiditis, and patients with diffuse multifocal carcinoma from those with a colloid or adenomatous goiter.
Several types of thyroid nodules can be reduced in size and function, even destroyed, using fine-needle techniques. The most straightforward is aspiration of fluid or blood from a purely cystic or hemorrhagic nodule, which disappears and does not recur. However, about half of cystic thyroid nodules do recur each time that fluid is removed (63). Some patients in whom the cysts recur choose surgery, and some physicians recommend surgery for patients whose cysts reappear after two or three aspirations, but repeated aspirations may lead eventually to disappearance of the cyst (63). The solid portion of nodules that are largely cystic should be examined via biopsy, with ultrasound guidance, ideally when the cyst is first aspirated.
Sclerosing agents, such as tetracycline or ethanol, have been used to prevent recurrence of cystic thyroid nodules (63,64), although tetracycline may be no more effective than saline (65), and extravasation of the sclerosing agent may damage the recurrent laryngeal nerve. Several groups have been trying to destroy solid, autonomously functioning thyroid nodules, or hypofunctioning nodules with benign cytology, by repeated injections of ethanol into the nodules under ultrasound guidance, with some success (64,66,67,68,69). The procedure can be quite painful, and only a small amount of ethanol can be injected at a time, and therefore multiple injections are usually needed. The long-term efficacy and safety of this procedure remain to be established, and the cost may be considerable because of the need for repeated injections. A more recent innovation is the use of laser photocoagulation to destroy benign functioning or nonfunctioning nodules (70,71). The clinical effectiveness, safety, and cost of this technique have not yet been established.
The presence of crystal clear, colorless, watery fluid is diagnostic of a parathyroid cyst, and measurement of parathyroid hormone in the cystic fluid will confirm the diagnosis. Serum calcium concentrations are usually normal. The cyst can be drained, but may recur.
Fine-needle biopsy can be used in patients who have had surgery for thyroid carcinoma and who have cervical lymphadenopathy or possible recurrence of tumor in the bed of the thyroid gland. If the nodes are palpable, they can be examined via biopsy in the usual fashion. More often, the abnormal nodes are detected by ultrasonography, and biopsy must be ultrasound-guided. In clinics with extensive ultrasound experience, the yield of diagnostically useful material in these patients is high (72,73), and ethanol injection to destroy isolated foci of residual thyroid cancer has recently been reported (73).
If the rare entity of acute suppurative thyroiditis is suspected, fine-needle aspiration can provide material for Gram stain and culture, and special stains have identified as the cause of both painful and painless thyroid enlargement in patients with human immunodeficiency virus infection (74).Pneumocystis carinii
MANAGEMENT OF PATIENTS WITH BENIGN THYROID NODULES
Patients Who Do Not Undergo Surgery
Most patients who present with a thyroid nodule prove to have a benign nodule, receive a recommendation against surgery, and accept the recommendation. For this very large group, the management options are observation or thyroxine (T4) therapy (ethanol injection or laser photocoagulation are other options currently available in a few centers). It is important to note that little is known about the natural history of benign thyroid nodules, although it is becoming clear that these nodules grow slowly in many, if not the majority of, patients (75,76,77).
For years, patients with benign thyroid nodules have been treated with T4, in the hope that decreasing TSH secretion would result in a decrease in nodule size, or at least prevent further nodule growth. The results of some uncontrolled studies supported the benefit of T4 therapy (78). Then, a series of controlled trials, most with a placebo group and all with repeated measurements of nodule volume by ultrasonography, found no effect or only modest benefit of T4 therapy in reducing nodule volume, usually defined as a 50% or greater decrease in volume (79,80,81,82,83). Despite these results, in a survey published in 2000, a substantial minority of thyroidologists in the United States and Europe said that they still recommended T4 therapy for selected patients with nodules (4). The main reasons for continued use of T4 therapy are some patients' desire for treatment, even if the likelihood of benefit is small; the hope for decreasing the number of patients who ultimately have surgery for benign nodules, because nodule growth causes cosmetic or pressure symptoms or fear of cancer; the hope for decreasing the development of additional nodules; and a possibility of cost saving by avoiding repetitive imaging studies and biopsies for patients with stable or growing nodules whose physicians do not consider benign needle biopsy results sufficiently reassuring. T4 treatment is ineffective in preventing the recurrence of cystic nodules (84). There is no reason to suppose that T4 treatment would be effective in autonomously functioning nodules.
There have been four meta-analyses of controlled trials of T4 therapy in patients with benign thyroid nodules, in which the number of nodules that decreased in volume by 50% or more was the measure of response (78,80,81,82). Each meta-analysis included at least one study that the others excluded for methodologic weaknesses or that was published afterward. In all these analyses the point estimate of relative likelihood of nodule shrinkage was in the 1.8 to 2.0 range in the T4-treated patients, as compared with the control patients. Whether the higher response rate was statistically significant depended on the statistical model used. Thus, even meta-analysis has been inconclusive. In pooled data from six controlled studies of 11 to 18 months' duration (46,82,83,85,86,87), with ultrasound measurements of nodule volume, 52 of 210 nodules (25) decreased in volume by more than 50% in the T4 treatment groups, as compared with 22 of 205 (11) in the control groups. Analysis of nodule growth in one meta-analysis revealed that T4 therapy for 6 to 12 months was associated with a lower frequency of a 50% or more increase in nodule volume, as compared with no treatment (8% vs. 17) (82).
Several individual trials of T4 treatment that used other response measures, and, therefore, could not be combined with other studies, found statistically significant responses to T4 treatment. The only controlled long-term study (75) followed patients with 1- to 3-cm solitary nodules for 5 years and used a least-measurable size difference (12% change) end point. In this study, 20 of 42 nodules (48) in the T4-treated patients decreased in size, as compared with 9 of 41 nodules (22) in the control patients; 12 nodules (29) enlarged in the T4-treated patients, as compared with 24 nodules (56) in the control patients; and three new nodules greater than 1 cm developed in the T4-treated patients, as compared with 12 new nodules in the control patients ( < 0.05, for all three comparisons). Two other uncontrolled observational studies using ultrasound measurements (76,77) similarly found that more than 50% of untreated nodules grew slowly. In the 5-year study (75) and two others, which were 12- and 18-month controlled trials of therapy (83,88), mean nodule volumes were significantly lower in the Tp4treated patients than in those given placebo or no treatment. In one 12-month study (85), which included some predominantly cystic nodules, there was no difference in mean nodule volume. One study with a crossover design found a significantly lower mean nodule volume in patients treated with T4 in doses that reduced serum TSH concentrations to low normal (0.4–0.6 mIU/L) for a year, as compared with a placebo period in the same patients (88). This approach is used by some American thyroidologists (37), and eliminates the potential for thyrotoxic symptoms and adverse skeletal and cardiac effects of suppressive doses of T4.
Overall, the possibility that some patients might benefit from T4 treatment has not been established, but has not yet been disproved either. Observation without treatment is currently the evidence-based choice.
More long-term studies of T4 therapy in patients with benign thyroid nodules are needed. The studies should assess not only changes in nodule volume, but also the ultrasound characteristics of the nodules; the appearance of new nodules; clinical outcomes, such as the number of patients who come to surgery or have cosmetic or compressive symptoms; and patient satisfaction. The relationship of serum TSH concentrations to outcome should also be assessed.
For all patients with a thyroid nodule, whether treated with T4 or not, nodule size should be reevaluated at regular intervals, at least annually. If T4 is given, the patient should have a measurement of serum TSH in 6 to 12 weeks, and periodically thereafter, to be sure that TSH secretion is not very low. Any patient in whom the nodule grows should have a repeat biopsy. If the biopsy again reveals benign thyroid cells in a patient who had not been treated, T
4 therapy can be considered then. If T4 was given at the outset, yet the nodule enlarged, continuing T4 is not a necessity, but is an option, since the nodule may have enlarged more had T4 not been given. Whenever T4 is given, the patient should be informed that the nodule may grow, and if the goal of therapy is a low serum TSH concentration, the risks of bone loss or cardiac arrhythmia should be discussed (see Chapter 40 and see Chapter 79).
Treatment after Thyroid Surgery
Patients who undergo thyroid lobectomy for what proves to be a benign thyroid nodule need no further therapy. These patients may have a small increase in serum TSH concentrations several months after surgery, but the increase is not sustained. Normal thyroid function should be verified by measurements of serum TSH 4 to 6 months after surgery. In the past, T4 was given routinely to these patients, but a few controlled studies have provided no evidence that the therapy prevents the formation of new nodules. An exception is patients who had head or neck irradiation and a partial thyroidectomy for a benign nodule, in whom T4 therapy does prevent new nodules (89). T4 therapy should prevent compensatory enlargement of the contralateral thyroid lobe, and is safe so long as TSH secretion is not suppressed to below the normal range.
SUMMARY OF MANAGEMENT OF PATIENTS WITH THYROID NODULES
Figure 73.2 shows a flow diagram of management steps and options for patients who present with a solitary thyroid nodule. Although the many possibilities make it a complex scheme, the simplest pathway applies to the most common group of patients. They are patients who have a normal serum TSH concentration and a biopsy that revealed only benign thyroid follicular cells, and for whom observation is the most appropriate recommendation.
FIGURE 73.2. Scheme for evaluation and treatment of patients with solitary thyroid nodules.
THYROGLOSSAL DUCT CYSTS
Thyroglossal duct cysts are collections of fluid in a persistent thyroglossal duct, the tract that forms as the thyroid gland descends from the base of the tongue to the lower neck during development (see Chapter 2). Most thyroglossal ducts disappear before birth, but they can persist and occasionally develop cysts, which are usually in the midline of the anterior neck, between the hyoid bone and the thyroid isthmus. Most are asymptomatic and discovered during childhood or adolescence, but some are discovered later in life. If fluid is aspirated, it usually does not contain thyroid follicular cells. Because these cysts can become infected, they should usually be removed surgically. Rare cases of thyroid carcinoma have been reported in thyroglossal duct cysts (90).
1. Mazzaferri EL. Management of a solitary thyroid nodule. 1993;328:553.N Engl J Med
2. Tan GH, Gharib H, Reading CC. Solitary thyroid nodule: comparison between palpation and ultrasonography. 1995;115:2418.Arch Intern Med
3. Marqusee E, Benson CB, Frates MC, et al. Usefulness of ultrasonography in the management of nodular thyroid disease. 2000;133:696.Ann Intern Med
4. Hegedüs L, Bonnema SJ, Bennedbaek FN. Management of simple nodular goiter: current status and future perspectives. 2003;24:102.Endocr Rev
5. Hamburger JI. Consistency of sequential needle biopsy findings for thyroid nodules. 1987;147:97.Arch Intern Med
6. Belfiore A, LaRosa GL, LaPorta GA, et al. Cancer risk in patients with cold thyroid nodules: relevance of iodine intake, sex, age, and multinodularity. 1992;93:363.Am J Med
7. Schlinkert RT, van Heerden JA, Goellner JR, et al. Factors that predict malignant thyroid lesions when fine-needle aspiration is “suspicious for follicular neoplasm.” 1997;72:913.Mayo Clin Proc
8. Hamming JF, Goslings BM, van Steenis GJ, et al. The value of fine-needle aspiration biopsy in patients with nodular thyroid disease divided into groups of suspicion of malignant neoplasms on clinical grounds. 1990;150:113.Arch Intern Med
9. Tuttle RM, Lemar H, Burch HB. Clinical features associated with an increased risk of thyroid malignancy in patients with follicular neoplasia by fine-needle aspiration. 1998;8:377.Thyroid
10. Raber W, Kaserer K, Niederle B, et al. Risk factors for malignancy of thyroid nodules initially identified as follicular neoplasia by fine-needle aspiration: results of a prospective study of one hundred twenty patients. 2000;10:709.Thyroid
11. Chen H, Nicol TL, Zeiger MA, et al. Hürthle cell neoplams of the thyroid: are there factors predictive of malignancy? 1998;227:542.Ann Surg
12. Elisei R, Bottici V, Luchetti F, et al. Impact of routine measurement of serum calcitonin on the diagnosis and outcome of medullary thyroid cancer: experience in 10,684 patients with nodular thyroid disorders. 2004;89:163.J Clin Endocrinol Metab
13. Karanikas G, Moameni A, Poetai C, et al. Frequency and relevance of elevated calcitonin levels in patients with neoplastic and nonneoplastic thyroid disease and in healthy subjects. 2004;89:515.J Clin Endocrinol Metab
14. Hamburger JI. The autonomously functioning thyroid nodule: Goetsch's disease. 1987;8:439.Endocr Rev
15. Farfel Z, Bourne HR, Iiri T. The expanding spectrum of G protein diseases. 1999;340:1012.N Engl J Med
16. Porcellini A, Ciullo I, Laviola L, et al. Novel mutations of thyrotropin receptor gene in thyroid hyperfunctioning adenomas. 1994;79:335.J Clin Endocrinol Metab
17. Liel Y. The yield of adequate and conclusive fine-needle aspiration results in thyroid nodules is uniform across functional and goiter types. 1999;9:25.Thyroid
18. Hegedüs L. Thyroid ultrasound. 2001;30:339.Endocrinol Metab Clin North Am
19. Gallo M, Pesenti M, Valcavi R. Ultrasound thyroid nodule measurements: the “gold standard” and its limitations in clinical decision making. 2003;9:194.Endocr Pract
20. Zarnegar R, Brunaud L, Clark OH. Prevention, evaluation, and management of complications following thyroidectomy for thyroid carcinoma. 2003;32:483.Endocrinol Metab Clin North Am
21. Falk SA. Complications of thyroid surgery: an overview. In: Falk SA, ed. , 2nd ed. Philadelphia: Lippincott-Raven, 1997:697.Thyroid disease: endocrinology, surgery, nuclear medicine, and radiotherapy
22. LoGerfo P. Coarse needle biopsy of the thyroid. In: Hamburger JI, ed. New York: Springer-Verlag, 1989:205.Diagnostic methods in clinical thyroidology.
23. Singer PA, Cooper DS, Daniels GH, et al. Treatment guidelines for patients with thyroid nodules and well differentiated thyroid cancer. 1996;156:2165.Arch Intern Med
24. Feld S, Garcia M, Baskin HJ, et al. AACE clinical practice guidelines for the diagnosis and management of thyroid nodules. 1996;2:78.Endocr Pract
25. Tan GH, Gharib H. Thyroid incidentalomas: management approaches to nonpalpable nodules discovered incidentally on thyroid imaging. 1997;126:226.Ann Intern Med
26. Ezzat S, Sarti DA, Cain DR, et al. Thyroid incidentalomas: prevalence by palpation and ultrasonography. 1994;154:1838.Arch Intern Med
27. Jemal A, Tiwari RC, Ghafoor A, et al. Cancer statistics 2004. 2004;54:8.CA Cancer J Clin
28. Schneider AB, Bekerman C, Leland J, et al. Thyroid nodules in the follow-up of irradiated individuals: comparison of thyroid ultrasound with scanning and palpation. 1997;82:4020.J Clin Endocrinol Metab
29. Leinung MC, Gianoukakis A, Lee DW. Ultrasonography in management of nodular thyroid disease. 2001; 135:383.Ann Intern Med
30. Leenhardt L, Hejblum B, Franc B, et al. Indications and limits of ultrasound-guided cytology in the management of nonpalpable thyroid nodules. 1999;84:24.J Clin Endocrinol Metab
31. Hagag P, Strauss S, Weiss M. Role of ultrasound-guided fine-needle aspiration biopsy in evaluation of nonpalpable thyroid nodules. 1998;8:989.Thyroid
32. Khurana KK, Richards VI, Chopra PS, et al. The role of ultrasonography-guided fine-needle aspiration biopsy in the management of nonpalpable and palpable thyroid nodules. 1998;8:511.Thyroid
33. Hatipoglu BA, Gierlowski T, Shore-Freedman E, et al. Fine-needle aspiration of thyroid nodules in radiation-exposed patients. 2000;10:63.Thyroid
34. Hamburger JI, Kaplan MM, Husain M. Diagnosis of thyroid nodules by needle biopsy. In: Braverman LE, Utiger RD, eds. , 6th ed. Philadelphia: JB Lippincott, 1991:544.Werner and Ingbar's the thyroid. A fundamental and clinical text
35. Tan GT, Gharib H, Goellner JR, et al. Management of thyroid nodules in pregnancy. 1996;156:2317.Arch Intern Med
36. Rosen IB, Korman M, Walfish PG. Thyroid nodular disease in pregnancy: current diagnosis and management. 1997;40:81.Clin Obstet Gynecol
37. Bennedbaek FN, Hegedüs L. Management of the solitary thyroid nodule: results of a North American survey. 2000;85:2493.J Clin Endocrinol Metab
38. Danese D, Sciacchitano S, Farsetti A, et al. Diagnostic accuracy of conventional versus sonography-guided fine-needle aspiration biopsy of thyroid nodules. 1998;8:15.Thyroid
39. Santos JEC, Leiman G. Nonaspiration fine needle cytology. 1988;32:353.Acta Cytol
40. Hales MS, Hsu, FSF. Needle tract implantation of papillary carcinoma of the thyroid following aspiration biopsy. 1990;34:801.Acta Cytol
41. Lever EG, Refetoff S, Scherberg NH, et al. The influence of percutaneous fine needle aspiration on serum thyroglobulin. 1983;56:26.J Clin Endocrinol Metab
42. Guidelines of the Papanicolaou Society of Cytopathology for the examination of fine-needle aspiration specimens from thyroid nodules. The Papanicolaou Society of Cytopathology Task Force on Standards of Practice. 1996;15:84.Diagn Cytopathol
43. Hamburger JI, Husain M, Nishiyama R, et al. Increasing the accuracy of fine-needle biopsy for thyroid nodules. 1989;113:1035.Arch Pathol Lab Med
44. Alexander EK, Herring JP, Benson CB, et al. Assessment of nondiagnostic ultrasound-guided fine needle aspirations of thyroid nodules. 2002;87:4924.J Clin Endocrinol Metab
45. Gharib H, Goellner JR. Fine-needle aspiration biopsy of the thyroid: an appraisal. 1993;118:282.Ann Intern Med
46. La Rosa GL, Belfiore A, Giuffrida D, et al. Evaluation of the fine needle aspiration biopsy in the preoperative selection of cold thyroid nodules. 1991;67:2137.Cancer
47. Gharib H. Changing concepts in the diagnosis and management of thyroid nodules. 1997;26: 777.Endocrinol Metab Clin North Am
48. Cochand-Priollet B, Guillausseau P-J, Chagnon S, et al. The diagnostic value of fine-needle aspiration biopsy under ultrasonography in nonfunctional thyroid nodules: a prospective study comparing cytologic and histologic findings. 1994;97:152.Am J Med
49. Carmeci C, Jeffrey RB, McDougall IR, et al. Ultrasound-guided fine-needle aspiration biopsy of thyroid masses. 1998; 8:283.Thyroid
50. Yang GCH, Liebeskind D, Messina AV. Ultrasound-guided fine-needle aspiration of the thyroid assessed by ultrafast Papanicolaou stain: data from 1135 biopsies with a two- to six-year follow-up. 2001;11:581.Thyroid
51. Sanchez RB, vanSonnenberg E, D'Agostino HB, et al. Ultrasound guided biopsy of nonpalpable and difficult to palpate thyroid masses. 1994;178:33.J Am Coll Surg
52. LiVolsi VA. Philadelphia: WB Saunders, 1990.Surgical pathology of the thyroid.
53. Raab SS, Silverman JF, Elsheikh TM, et al. Pediatric thyroid nodules: disease demographics and clinical management as determined by fine needle aspiration biopsy. 1995; 95:46.Pediatrics
54. Degnan BM, McClellan DR, Francis GL. An analysis of fine-needle aspiration biopsy of the thyroid in children and adolescents. 1996;31:903.J Pediatr Surg
55. Chehade JM, Silverberg AB, Kim J, et al. Role of repeated fine-needle aspiration of thyroid nodules with benign cytologic features. 2001;7:237.Endocr Pract
56. Lucas A, Llatjos M, Salinas I, et al. Fine-needle aspiration cytology of benign nodular thyroid disease. Value of re-aspiration. 1995;132:677.Eur J Endocrinol
57. Erdogan MF, Kamel N, Aras D, et al. Value of re-aspirations in benign nodular thyroid disease. 1998;8:1087.Thyroid
58. Cersosimo E, Gharib H, Suman VJ, et al. “Suspicious” thyroid cytologic findings: outcome in patients without immediate surgical treatment. 1993;68:343.Mayo Clin Proc
59. McHenry CR, Walfish PG, Rosen IB. Non-diagnostic fine needle aspiration biopsy. A dilemma in management of thyroid disease. 1993;59;415.Am Surgeon
60. Baloch ZW, Sack MJ, Yu GH, et al. Fine-needle aspiration of thyroid: an institutional experience. 1998;8:565.Thyroid
61. Cheng H-Y, Lin J-D, Chen J-F, et al. Correlation of fine needle aspiration cytology and frozen section biopsies in the diagnosis of thyroid nodules. 1997;50:1005.J Clin Pathol
62. Paphavasit P, Thompson GB, Hay ID, et al. Follicular and Hürthle cell thyroid neoplasms. Is frozen section evaluation worthwhile? 1997;132:674.Arch Surg
63. Lee J-K, Tai F-T, Lin H-D, et al. Treatment of recurrent thyroid cysts by injection of tetracycline or minocycline. 1989;149:599.Arch Intern Med
64. Zingrillo M, Torlontano M, Ghiggi MR, et al. Percutaneous ethanol injection of large thyroid cystic nodules. 1996; 6:403.Thyroid
65. Hegedüs L, Hansen JM, Karstrup S, et al. Tetracycline for sclerosis of thyroid cysts. A randomized study. 1988;148:1116.Arch Intern Med
66. Pacini F. Role of percutaneous ethanol injection in management of nodular lesions of the thyroid gland. 2003;44: 211.J Nucl Med
67. Zingrillo M, Collura D, Ghiggi MR, et al. Treatment of large cold benign thyroid nodules not eligible for surgery with percutaneous ethanol injection. 1998;83: 3905.J Clin Endocrinol Metab
68. Bennedbaek FN, Nielsen LK, Hegedüs L. Effect of percutaneous ethanol injection therapy versus suppressive doses of l-thyroxine on benign solitary solid cold thyroid nodules: a randomized trial. 1998;83:830.J Clin Endocrinol Metab
69. Caraccio N, Goletti O, Lippolis PV, et al. Is percutaneous ethanol injection a useful alternative for the treatment of the cold benign thyroid nodule? Five years' experience. 1997;7:699.Thyroid
70. Dossing H, Bennedbaek FN, Karstrup S, et al. Benign solitary solid cold thyroid nodules: US-guided interstitial laser photocoagulation—initial experience. 2002;225:53.Radiology
71. Dossing H, Bennedbaek FN, Hegedüs L. Ultrasound-guided interstitial photocoagulation of an autonomous thyroid nodule: the introduction of a novel alternative. 2003;13:885.Thyroid
72. Frasoldati A, Pesenti M, Gallo M, et al. Diagnosis of neck recurrences in patients with differentiated thyroid carcinoma. 2003;97:90.Cancer
73. Lewis BD, Hay ID, Charboneau JW, et al. Percutaneous ethanol injection for treatment of cervical lymph node metastases in patients with papillary thyroid carcinoma. 2002;178:699.AJR
74. Guttler R, Singer PA. thyroiditis: report of three cases and review of the literature. 1993; 153:393.Pneumocystis cariniiArch Intern Med
75. Papini E, Petrucci L, Guglielmi R, et al. Long-term changes in nodular goiter: a 5-year prospective randomized trial of levothyroxine suppressive therapy for benign cold thyroid nodules. 1998;83:780.J Clin Endocrinol Metab
76. Alexander EK, Hurwitz S, Heering JP, et al. Natural history of benign solid and cystic thyroid nodules. 2003; 138:315.Ann Intern Med
77. Quadbeck B, Pruellage J, Roggenbuck U, et al. Long-term follow-up of thyroid nodule growth. 2002;110:348.Exp Clin Endocrinol Diabetes
78. Richter B, Nesies G, Clar C. Pharmacotherapy for thyroid nodules: a systematic review and meta-analysis. 2002;31:699.Endocrinol Metab Clin North Am
79. Gharib H, Mazzaferri EL. Thyroxine suppressive therapy in patients with thyroid nodular disease. 1998;128: 386.Ann Intern Med
80. Csako G, Byrd D, Wesley RA, et al. Assessing the effects of thyroid suppression on benign solitary thyroid nodules: a model for using quantitative research synthesis. 2000;79:9.Medicine
81. Castro MR, Caraballo PJ, Morris JC. Effectiveness of thyroid hormone suppressive therapy in benign solitary thyroid nodules: a meta-analysis. 2002;87:4154.J Clin Endocrinol Metab
82. Zelmanovitz F, Genro S, Gross JL. Suppressive therapy with levothyroxine for solitary thyroid nodules: a double-blind controlled clinical study and cumulative meta-analyses. 1998;83:3881.J Clin Endocrinol Metab
83. Wemeau J-L, Caron P, Schvartz C, et al. Effects of thyroid—stimulating hormone suppression with levothyroxine in reducing the volume of solitary thyroid nodules and improving extranodular nonpalpable changes: a randomized, double-blind, placebo-controlled trial by the French thyroid group. 2002;87:4928.J Clin Endocrinol Metab
84. McCowen K, Reed J, Fariss B. The role of thyroid therapy in patients with thyroid cysts. 1980;68:853.Am J Med
85. Larijani B, Pajouhi M, Bastanhagh MH, et al. Evaluation of suppressive therapy for cold thyroid nodules with levothyroxine: double-blind placebo-controlled clinical trial. 1999;5:251.Endocr Pract
86. Reverter J, Lucas A, Salinas I, et al. Suppressive therapy with levothyroxine for solitary thyroid nodules. 1992;36:25.Clin Endocrinol (Oxf)
87. Papini E, Bacci V, Panunzi C, et al. A prospective randomized trial of levothyroxine suppressive therapy for solitary thyroid nodules. 1993;38:507.Clin Endocrinol (Oxf)
88. Koc M, Ersoz HO, Akpinar I, et al. Effect of low- and high-dose levothyroxine on thyroid nodule volume: a crossover placebo-controlled trial. 2002;57:621.Clin Endocrinol (Oxf)
89. Fogelfeld L, Wiviott MBT, Shore-Freedman E, et al. Recurrence of thyroid nodules after surgical removal in patients irradiated in childhood for benign conditions. 1989;320: 835.N Engl J Med
90. Doshi SV, Cruz RM, Hilsinger RL Jr. Thyroglossal duct carcinoma: a large case series. 2001;110: 734.