| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Editorial |
Chief, Endocrine Section, Washington Hospital Center, Washington, D.C. 20010
Address all correspondence and requests for reprints to: Kenneth D. Burman, Chief, Endocrine Section, Washington Hospital Center, 110 Irving Street N.W., Washington, D.C. 20010. E-mail: Kenneth.D. Burman{at}MedStar.net.
It is controversial whether microscopic papillary thyroid cancer (defined as being equal to or less than 1.0 cm) has a significant risk of local or distant metastasis or recurrence and how patients with thyroid nodules less than 1–1.5 cm should be approached medically and surgically (1, 2, 3). In the current issue of JCEM, Roti et al. (4) present important and controversial data regarding microscopic papillary thyroid cancer. They retrospectively analyzed 243 patients with microscopic papillary thyroid cancer representing a surprising 39% of all patients seen with thyroid cancer from 1993–2002. They apparently routinely aspirate nodules 5 mm or larger. The authors did not comment on how many aspirations they performed in total and what percentage represented a benign aspirate. Fifty-two subjects had an unsuspected or incidental finding of microscopic papillary thyroid cancer detected on the histological sample in a patient operated on for Graves’ disease or presumed benign nodular goiter; 191 subjects had a thyroid nodule where the presence of papillary cancer was suspected (not incidental). Because 92% of the patients underwent a total thyroidectomy and 96% were subsequently treated with 131-I (50–100 mCi), this study cannot furnish controlled information regarding the potential benefit (or harm) of these procedures. Similarly, because all patients with cancer received TSH suppressive doses of levothyroxine, the role of TSH suppression in affecting disease recurrence also cannot be assessed, and lastly, long-term morbidity or mortality cannot be determined.
The patients with microscopic papillary thyroid cancer were followed for a median of 5.1 yr, and they had standard monitoring techniques including clinical examinations, serum thyroid function tests, thyroglobulin levels, neck sonograms, and radioiodine scans on a periodic basis. There were no statistical clinical differences between the patients with incidentally discovered or suspected microscopic papillary thyroid cancer in terms of tumor size, multifocality, bilaterally, or frequency of local extension. However, caution should be displayed in assessing these numbers because the frequency of bilaterality was 11 vs. 20%, multifocality 19 vs. 36%, and lymph node metastases 4 vs. 16% in the incidental group compared with the nonincidental group, respectively. The lack of statistical significance could be a reflection of the numbers of patients involved, and further studies addressing these issues are warranted. Based on the findings noted, there appears to be a trend that nonincidental microscopic papillary thyroid cancer is more aggressive (4). It would be interesting to know what the preoperative fine needle aspiration diagnosis was in the patients with a known thyroid nodule. A striking finding is that familial papillary thyroid cancer was approximately twice as common in the incidental group (9.6 vs. 4.2%). The mean diameter of the papillary thyroid cancers was about 0.55–0.56 cm (range, 0.1–1.0 cm) in both the incidental and nonincidental groups.
There are several important broad issues to consider. There is a high frequency of thyroid nodules less than 1 cm in the general population, and a significant number of individuals who die of other reasons have thyroid cancers incidentally identified at autopsy that apparently did not adversely affect the patients’ health (5). Specifically, in an autopsy survey, 50.5% of the glands did not contain a nodule, 12.2% contained a single nodule, 37.3% contained multiple nodules, and 4.2% of the nodular glands (2.1% of all glands) contained thyroid cancer (5). In fact, 67% of the population may have thyroid nodules (22% solitary, 45% multiple) when assessed by thyroid sonogram (6). On the other hand, clinically relevant thyroid cancer with local and even distant metastases or recurrences can occur in nodules less than 1 cm (7, 8). How do we integrate these opposing factors when confronted with a patient with a nodule less than 1 cm? Although the Society of Radiologists in Ultrasound Consensus Statement suggests that the importance of an early diagnosis of thyroid cancer is uncertain (9), many authorities believe it is important to diagnose thyroid cancer at an early stage because this may reduce the risk of recurrence, and possibly mortality (10, 11). In the study of Roti et al. (4) 79% of the patients with microcarcinoma had classic papillary thyroid cancer, and it seems that a relatively low number had the follicular variant of papillary thyroid cancer (13%). The presence of a more problematic type of thyroid cancer, such as tall cell variant, was rare, and these potentially more aggressive types were not responsible for the findings of local invasion or spread.
In sum, the most important aspects of the study by Roti et al. (4) are that: 1) tumors equal to or less than 1 cm accounted for 39% of all cancers; 2) local node invasion occurred in 0.8% of tumors less than 5 mm and in 1.6–3.3% of patients with nodules 6–10 mm in size; 3) distant metastases and recurrence occurred in about 0.4–0.8% of patients with 8–10 mm size tumors; and 4) only tumors 8 mm or larger were associated with distant metastases and recurrence. Are these findings by Roti et al. (4) consonant with recently published consensus guidelines (12)? A task force of thyroidologists [sponsored by the American Thyroid Association (ATA)] concluded: "Generally, only nodules larger than 1 cm should be evaluated because they have the potential to be clinically significant cancer. Occasionally, there may be nodules smaller than 1 cm that require evaluation, because of suspicious ultrasound findings, a history of head and neck irradiation, or a positive family history of thyroid cancer" (12). The American Association of Clinical Endocrinologists (AACE) (and Associazione Medici Endocrinologi) guidelines indicate that "an arbitrary diameter cutoff of 10 or 15 mm for cancer risk is not justified" (13). The Society of Radiology guidelines suggest considering a fine needle aspiration in nodules 2 cm or greater (9). If a nodule is at least 1 cm with microcalcifications on sonogram or at least 1.5 cm and is mainly solid or contains coarse calcification, they would also then strongly consider a fine needle aspiration. They observe that the presence of microcalcifications, hypoechogenicity, irregular margins, mainly solid components, enhanced intranodular vascularity, and the nodule being taller rather than wider are each sonographic features associated with a higher likelihood of cancer (9).
Roti et al. (4) do not divide their patients based on TMN staging, fine needle aspiration results, or whether the sonogram showed abnormal or suspicious cervical nodes before surgery. It appears that 13 of Roti’s patients who had local or distant metastases (out of 34 patients) were under age 45, and an additional four patients were between ages 45 and 50 yr. The patients under age 45 (n = 13) would have had stage 1 disease, and the patients 45 yr or older would likely have had stage 3 disease. If the ATA (12) and Society of Radiology guidelines (9) would have been followed (assuming there were no suspicious findings on history, physical examination, or radiological findings), it is possible that none of the 191 patients with nonincidental papillary thyroid cancer would have been detected, at least initially, because they would not have had a fine needle aspiration. In the nonincidental group, 34 patients (18% of those with cancer) with extrathyroidal extension, 30 (16%) with lymph node diagnosis, and four (2%) with distant metastases presumably also would not be initially detected. The precise number of patients not detected would depend on the clinical context and radiological findings.
Kim et al. (14) assessed patients with 155 solid nodules and observed that size of the nodule (0.3–2.8 cm; mean, 0.75 cm) was not predictive of malignancy. Chow et al. (2) and Pelizzo et al. (15) also did not find a difference in the frequency of local node metastases or aggressiveness in tumors greater than and less than 5 mm in size. Still, in the study of Roti et al. (4), 13% of all patients with a thyroid cancer less than 1 cm had local node metastases, 1.7% had recurrent or persistent disease, and there was no mortality (remembering the patients were treated with total thyroidectomy and 131-I therapy). Only four patients had distant metastases (two in mediastinum and two in lung). Having node metastases, in general, is thought to increase the risk of recurrence and patient morbidity, even if mortality appears unchanged (7, 11, 16, 17, 18).
Regardless of whether a small nodule is aspirated, it seems important to monitor all nodule patients with periodic sonograms to determine whether the nodule is growing or changing its characteristics. There are limitations, however, to using sonography to monitor nodules, and the technique may not detect small changes. Roti’s (4) study did not give all three dimensions of these nodules, so volume assessments could not be made. The ATA guidelines indicate that "... it is recommended that serial ultrasound be used in follow-up of thyroid nodules to detect clinically significant changes in size. There is no consensus on the definition of nodule growth, however, or the threshold that would require biopsy" (12). Cooper et al. (12) suggested that criteria are either a 15% increase in nodule volume or mean diameter or a 20% increase in nodule diameter (associated with a minimum increase of at least 2 mm in two dimensions). However, in a study of only 14 patients with 27 nodules, Asanuma et al. (19) suggested that the rate of nodule growth as assessed by serial sonograms could not distinguish benign from malignant thyroid nodules. Nonetheless, the recommendation to monitor nodules serially by sonogram and consider reaspiration for those that grow seems reasonable. Implicit in these recommendations is the performance of sonography with modern equipment performed by knowledgeable, interested health caregivers with direct comparisons with previous images.
In the study by Roti et al. (4), multifocality occurred in 19 and 36% of patients, and bilaterality occurred in 11 and 20% of patients in the incidental and nonincidental groups, respectively (4). Considering both groups together, multifocality and the presence of cervical nodes containing cancer were observed in 32 and 13% of patients, respectively, and there was not a correlation between the presence of multifocality, bilaterality, and the presence of cervical adenopathy (4). It is difficult for the clinician to know how to approach patients with multifocality or bilateral lesions (4). Roti’s article (4) does not make it clear whether all nodules were aspirated if present bilaterally or in multiple sites. The ATA guidelines do not address how to approach thyroid nodules less than 1 cm that are bilateral or multifocal (12). A previous paper had suggested that familial papillary cancer represented a significant increased risk for cervical and distant metastases (20) and for occurrence in first-degree relatives (21), but Roti’s study (4) did not find that familial papillary thyroid cancer was more aggressive. Patients with familial papillary thyroid cancer had comparable rates of multifocality, bilaterality, capsular invasion, and node metastasis as did patients with nonfamilial papillary thyroid cancer. Mazzaferri and Jhiang’s (22) oft-quoted analysis of multifocality showed a higher mortality rate with three or more foci. However, they considered papillary and follicular cancer together, and the precise size of the multifocal lesions is not presented.
Davies (23) notes that it is illogical to rely solely on size criteria to determine which nodules to aspirate. Baskin and Duick (10) also note that using sonographic criteria to decide which nodules may be malignant is relatively insensitive. For example, the presence of microcalcifications has only a 29% sensitivity (10, 24).
How can these wide ranging unanswered or controversial questions be considered to arrive at a reasonable approach when there are no long-term controlled data? Any suggested approach at least to some extent represents opinion. The critical issue is: Is there a clinical difference in morbidity or mortality if a microcarcinoma is detected when it is less than 1 cm compared with when it is larger than 1 cm? Hay et al. (25) studied 535 patients with microscopic papillary thyroid cancer over a mean of 16 yr, with a range up to 48 yr. The mortality rate was 0.4%, and the 20-yr recurrence rate was 6%. In 203 patients with micropapillary thyroid cancer, Chow et al. (26) found that cause-specific mortality was 1%, lymph node recurrence rate was 5%, and distant metastasis occurred in 2.5%. Mazzaferri and Jhiang (22) showed that for both papillary and follicular cancers, "there was a linear relationship between tumor size and recurrence and cancer mortality." This study gives some justification for using a tumor cutoff size because patients with tumors less than 1.5 cm had 30-yr mortality of 0.4%, with 1.5- to 4.4-cm tumors a 7% mortality and with tumors larger than 4.5 cm an 18% mortality (11, 22).
On the other hand, it must be noted that Ito et al. (27) followed 162 patients with biopsy-proven papillary thyroid cancers that were equal to or less than 10 mm. Seventy percent of these tumors either got smaller or did not change size over about 8 yr; 10.2% of nodules enlarged by more than 10 mm, and node metastases occurred in 1.2%. Of 626 patients with the same initial tumor characteristics and positive biopsy (570 originally in surgery group; 56 originally in observation group) who underwent a thyroidectomy, local node metastases were observed in 50.5% and multifocality in 42.8%, and recurrence occurred in 5% at 8 yr after surgery. The authors concluded that micropapillary thyroid cancer does not frequently change when assessed clinically, whereas histologically local nodal metastasis is common (27).
Given each of these factors, the ATA guidelines seem reasonable in suggesting that a fine needle aspiration be considered on all thyroid nodules 10 mm or larger. Nodules 8–10 mm in size should also be considered for fine-needle aspiration if the nodule has clinical or sonographic characteristics that are worrisome (e.g. microcalcifications, enhanced Doppler flow) (28) or occurs in a patient with a known family history of papillary thyroid cancer (21). Perhaps the co-occurrence of Graves’ disease and a thyroid nodule should also warrant a biopsy of the nodule, given the observation that circulating TSH receptor antibodies may increase the behavior of the thyroid cancer (29). A history of radiation exposure and head and neck external radiation or exposure should also trigger a fine needle aspiration. These comments mainly apply to solitary nodules 8–10 mm, and all solitary nodules larger than 10 mm should be routinely aspirated to the extent possible. If worrisome aspects apply, even smaller solitary nodules between 5 and 8 mm should be considered for aspiration on an individual basis, based on personal view, patient desires, clinical context, and sonographic appearance. It is expected that only a very small percentage of such nodules would actually be aspirated.
Approaching multinodular goiters, the ATA guidelines suggest: "In the presence of two or more thyroid nodules larger than 1–1.5 cm, those with a suspicious sonographic appearance should be aspirated preferentially." Further, "If none of the nodules has a suspicious sonographic appearance and multiple sonographically similar coalescent nodules are present, the likelihood of malignancy is low, and it is reasonable to aspirate the largest nodules only" (12). In a multinodular goiter, the AACE guidelines emphasize sonographic characteristics rather than nodule size in determining which nodule(s) to aspirate (13). In my view, it seems that elements of both guidelines are correct but that sonographic appearance alone has not been well studied especially in terms of consistency between observers. The AACE guidelines suggest that if a nodule has at least two sonographic criteria for malignancy, the sensitivity is between 87 and 93% (13). This is high, but it suggests to me that individual judgment and aspiration of the largest nodules seem appropriate, but if particular nodules have worrisome sonographic characteristics, regardless of size, they should also be considered for aspiration.
Irrespective of which suggested minimum measurement or guideline is used, the most critical aspect is to monitor the patient with periodic clinical examinations and sonograms. Using this approach, it is likely that an increase in nodule size, perhaps reflecting the presence of thyroid cancer, could be detected and the nodule aspirated. This approach is not perfect and will miss some patients with nodules smaller than 8–10 mm that already have local or even more rarely distant metastases, but until a more detailed cost-benefit analysis and long-term controlled studies are performed, this approach reflects our imperfect clinical views. The acquisition of more knowledge regarding genetic markers may allow better discrimination of nodules that will display more aggressive clinical behavior.
Received April 12, 2006.
Accepted April 21, 2006.
References
This article has been cited by other articles:
 |
|
 |
|
  C. R. McCartney and G. J. Stukenborg Decision Analysis of Discordant Thyroid Nodule Biopsy Guideline Criteria J. Clin. Endocrinol. Metab., August 1, 2008; 93(8): 3037 - 3044. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
