Fax: (416) 586-8481
Predictors of malignancy in thyroid fine-needle aspirates “cyst fluid only” cases
Can potential clues of malignancy be identified?
Article first published online: 26 AUG 2009
Copyright © 2009 American Cancer Society
Volume 117, Issue 5, pages 305–310, 25 October 2009
How to Cite
Jaragh, M., Carydis, V. B., MacMillan, C., Freeman, J. and Colgan, T. J. (2009), Predictors of malignancy in thyroid fine-needle aspirates “cyst fluid only” cases. Cancer Cytopathology, 117: 305–310. doi: 10.1002/cncy.20041
- Issue published online: 12 OCT 2009
- Article first published online: 26 AUG 2009
- Manuscript Accepted: 27 MAR 2009
- Manuscript Revised: 25 MAR 2009
- Manuscript Received: 26 JAN 2009
- Kuwait Institute for Medical Specialization, Kuwait
- thyroid fine-needle aspiration;
- papillary carcinoma
A recent consensus conference on thyroid fine-needle aspiration (FNA) cytology concluded that specimens with abundant histiocytes and few or no follicular cells should be interpreted as “cyst fluid only,” under the category of “nondiagnostic.” The purpose of the current study was to identify any cytomorphologic characteristics in this type of specimen that are predictive of an underlying malignancy.
Thyroid FNA cases with a report of cyst fluid only and a follow-up thyroidectomy specimen were identified during a 3-year period. A blinded retrospective review of 6 morphologic features in the thyroid FNA specimens was conducted. These review findings were then correlated with the histopathologic diagnosis (benign or malignant).
Of the 76 cyst fluid only cases with subsequent thyroidectomy, 10 cases had an ipsilateral diagnosis of papillary carcinoma measuring ≥1.0 cm. There was no association found between the number or amount of acute inflammatory cells, blood, colloid, macrophages, and pigmented macrophages and the histologic outcome. In only 4 of the 10 cases with a malignant outcome was the specimen assessed as being truly inadequate on retrospective review, and in 1 of these cases, the cytology was suggestive of malignancy.
The only cytomorphologic characteristic found to be predictive of subsequent malignancy in cyst fluid only cases was the presence of follicular epithelium with atypical or suspicious features. Therefore, cases containing atypical epithelial cells should not be categorized as nondiagnostic or cyst fluid only, but rather diagnosed as atypical or suspicious. Cancer (Cancer Cytopathol) 2009. © 2009 American Cancer Society.
A variety of hyperplastic and neoplastic benign and malignant thyroid nodules may undergo cystic degeneration characterized by the accumulation of fluid and numerous histiocytes and inflammatory cells with adjacent fibrosis. Typically, such cystic degeneration can be detected in fine-needle aspiration (FNA) specimens by the presence of numerous foamy macrophages, hemosiderin-laden macrophages, scant colloid, and rare follicular epithelial cells. The appropriate cytodiagnostic and adequacy categorization for these specimens has been controversial. One task force suggested that such findings be reported as “consistent with benign thyroid cyst.”1 The recent National Cancer Institute (NCI) Thyroid Fine Needle Aspiration State of the Science Conference concluded that thyroid FNA specimens with few to no follicular cells should be interpreted as ‘cyst fluid only,’ under the category of “nondiagnostic,” possibly with the additional citation of “limited,” and not “unsatisfactory.”2-4 The underlying risk of malignancy, including papillary carcinoma in particular, in simple noncomplex cysts is <5%, but rises with larger size (>3 cm), any associated complexity, and disease recurrence.2, 4 Consequently, an optional recommendation for correlation with the cyst size and complexity and a disclaimer that a cystic carcinoma cannot be entirely excluded may be added.2, 4 The purpose of the current study was to identify any cytomorphologic characteristics of these cyst fluids that are predictive of a subsequently detected malignancy (ie, a false-negative cytologic interpretation). If such characteristics could be identified, then further refinement of criteria for these types of specimens could be possible.
MATERIALS AND METHODS
In the Pathology and Laboratory Medicine Department of Mount Sinai Hospital, thyroid FNAs that demonstrate abundant histiocytes and no or inadequate numbers of follicular cell groupings have been interpreted as “degenerating lesion” for several years. The clinical response to a report of degenerating lesion varies with the consideration of other clinical and imaging data. This local cytodiagnostic category is equivalent to the recommended nondiagnostic category of “cyst fluid only“ of the NCI Thyroid Fine Needle Aspiration State of the Science Conference.1 In the Mount Sinai Hospital laboratory, approximately 7% of all thyroid FNAs are labeled as “degenerative lesion.” For the purposes of this study and report, and conformity to the National Institutes of Health (NIH) thyroid terminology, these cases are referred to as “cyst fluid only.”
After Mount Sinai Hospital Research Ethics Board approval, all cases of cyst fluid only with a subsequent hemithyroidectomy or total thyroidectomy specimen were identified in the archives from January 2005 to December 2007. The subsequent histopathologic diagnosis of the dominant nodule was accepted as the final or gold standard diagnosis for each case, including incidental small carcinomas, which were also recorded. For each case the patient's age, sex, and type of surgery and the sites of the lesion and FNA were recorded. At our institution, thyroid FNAs are performed by radiologists, surgeons, and endocrinologists without onsite assessment by any cytologist. The number of slides and types of available preparations, whether ThinPrep (Hologic, Marlborough, Mass), conventional clinical smears, or conventional laboratory smears were also noted. All preparations were stained using a modified Papanicolaou method in a Leica (Nussloch, Germany) autostainer.
A microscopic review of all FNA cases was undertaken blinded to the histopathologic outcome. The following cytomorphologic features were assessed: the presence or absence of follicular cells (none, <6 groups of at least 10 cells, 6-8 groups, and >8 groups), foamy macrophages (none = 0, <10 per 10 high-power fields = 1, 10-20 = 2, and >20 = 3), pigment-laden macrophages (none = 0, <10 per 10 high-power fields = 1, 10-20 = 2, and >20 = 3), colloid (little = 1, moderate amount = 2, and abundant = 3), blood (little = 1, moderate amount = 2, and abundant = 3), and polymorphonuclear inflammatory cells (none = 0, <10, 10-20, and >20 per 10 high power fields). If any follicular epithelium was present, architectural and cytologic features of thyroid papillary carcinoma (such as syncytial fragments, nuclear crowding, and overlapping, irregular nuclear membrane; micronucleoli; powdery chromatin; nuclear grooves; and nuclear pseudoinclusions) were evaluated.
The histopathologic outcomes of subsequent thyroidectomy specimens were classified as benign or malignant (papillary carcinoma, <1 cm, and ≥1 cm).
Seventy-six cases of cyst fluid only with subsequent histopathologic diagnoses were identified during the 3-year study period. These 76 cases represented <10% of all thyroid FNAs reported as cyst fluid only during the study period, and were reported by 6 pathologists. Of the 76 cases, 59 were from women, with a mean age of 50 years (range, 21-76 years), and 17 were from men, with a mean age of 58 years (range, 28-83 years). The distribution of the types of thyroid FNA preparations was as follows: 17 cases of ThinPrep only, 3 cases of clinical smears only, 41 cases of ThinPrep and clinical smears, 5 cases of ThinPrep and laboratory smears, and 10 cases of ThinPrep and laboratory and clinical smears. A mean of 2 slides were prepared in ThinPrep cases and laboratory-prepared smears. The number of clinic-prepared smears varied from 1 to 6 per case.
The final histopathologic diagnosis of the 76 cases of cyst fluid only is shown in Table 1. Fifty-one of the 76 cases of cyst fluid only (67.5%) demonstrated either a degenerating adenoma or colloid nodule on final histopathologic evaluation. An additional 5 cases demonstrated a benign lesion ipsilateral to the FNA, and an incidental carcinoma in the opposite lobe of a total thyroidectomy specimen, which was unsampled by the FNA.
|Malignant, %||Benign, %|
|≥1 cm||<1 cm|
In 10 cyst fluid only cases, the dominant nodule of interest was an ipsilateral papillary carcinoma, which measured ≥1.0 cm (largest, 4 cm) and consisted of the following histologic types: 5 follicular variants of papillary carcinoma, 3 classic papillary carcinoma, and 2 oncocytic variants of papillary carcinoma.
The remaining 10 cases demonstrated ipsilateral carcinomas measuring <0.5 cm. These small carcinomas almost certainly represented “incidental” microcarcinomas, and were accompanied by a dominant nodule, but the possibility of their sampling cannot be excluded. These results are reported separately from the categories of benign and carcinomas measuring ≥1 cm shown in Table 1.
The blinded microscopic review of the 76 cases was as follows. Variable amounts of colloid were present. Twenty-three cases demonstrated no evidence of colloid, 40 cases had 1+ colloid, and 13 cases demonstrated 2+ colloid. In contrast, foamy macrophages, 1 of the defining features of degenerating lesions, were evident in all cases, and were ≤2 in all cases. Similarly, pigmented macrophages, another defining feature, were also present in all cases with 1 exception, with 23 cases demonstrating <10 per 10 high-power fields, 27 cases demonstrating between 10 and 20 per 10 high-power fields, and 26 cases demonstrating >20 per 10 high-power fields. Blood was present in all the cases in variable amounts from 1 to 3+. Acute inflammatory cells in the background were fairly uniform, with 71 of 76 cases demonstrating the presence of these cells but consisting of <10%.
A comparison of the review findings for colloid, macrophages, pigmented macrophages, and blood for cases with benign outcome on the ipsilateral side, versus ipsilateral carcinoma ≥1 cm outcome, is shown in Figure 1. Only carcinomas measuring ≥1 cm were used for this comparison, because these represented the lesion of interest. (No comparison was made for the presence of acute inflammatory cells, because the review had produced a uniform finding throughout the 76 cases.) There was no association or correlation noted between any of these features (macrophages, pigmented macrophages, colloid, and blood) and subsequent histopathologic (malignant) outcome.
The presence or absence of follicular cells in both benign and malignant cases is shown in Table 2. The majority of cyst fluid only cases with a benign outcome (43 of 56) demonstrated no or inadequate numbers of follicular groups. None of the cases with follicular epithelium exhibited any atypia. In contrast, 5 of 10 cases with an ipsilateral malignancy measuring ≥1 cm did demonstrate ≥6 groups of at least 10 follicular cells, and were adequate. In 3 of these 5 cases, some cytologic features suggestive of papillary carcinoma (syncytial pattern, nuclear crowding, powdery chromatin, thick and thin nuclear rims, and micronucleoli) were present (Fig. 2), and another case exhibited prominent nuclear grooves, whereas the final case demonstrated no epithelial abnormalities. Four of the 10 cases of carcinoma measuring ≥1 cm had inadequate numbers of follicular groups; in 1 of these 4 cases some cytologic features suggestive of papillary carcinoma (ie, syncytial pattern, nuclear crowding, thick and thin nuclear rims, and micronucleoli) were present (Fig. 3). The remaining 10th case of carcinoma measuring ≥1 cm demonstrated no follicular cells and an abundance of pigmented macrophages only.
|Groups Based on No. of Follicular Cells|
|No Follicular Cells||<6||6-8||>8|
Four of the 10 carcinomas measuring <1 cm had adequate groups of follicular cells, and in 2 of these 4 cases some of the nuclear features of papillary carcinoma were observed. Of the remaining 6 cases of carcinomas measuring <1 cm, 2 had no follicular cells and 4 had less than adequate groups of unremarkable follicular cells.
The cases of cyst fluid only in the current study are not representative of the general population of patients with this cytologic interpretation. Only a small minority of cases interpreted as cyst fluid only during the 3-year period of the study were included in the study group, because cases were selected only if patients had proceeded to surgery. The decision to undertake surgery was based on several factors, including clinical and imaging findings, cyst recurrence after aspiration, FNA cytology, any repeat cytologic aspirate findings, surgeon's judgment, and patient's preference. Identifying which single factor was the key to the surgical decision-making process is not possible from available data. This patient group, therefore, could be considered to be a high-risk group for malignancy. Greater than 25% of the patients were found to have an ipsilateral thyroid carcinoma on final histopathologic examination.
This study found that a significant proportion of cases of cyst fluid only (22 of 76) should have been designated as adequate as determined by the retrospective review, using adequacy criteria proposed by the NCI Thyroid Fine Needle Aspiration State of the Science Conference. If strict criteria for the diagnosis of cyst fluid only cases, such as those outlined by the NIH Consensus Conference, are not used, then a proportion of this diagnostic category is and will continue to be false-negative FNAs. Notably, of the 11 malignant study cases that had inadequate numbers of follicular cell groups, a major NIH Consensus Conference criterion for cyst fluid only, only a single case demonstrated atypical cytologic features. The detection of any cytologic atypia, such as in this case, would also preclude the interpretation of inadequate/nondiagnostic. The results of the current study suggest that the cyst fluid only interpretive category may not be highly reproducible. To the best of our knowledge, there are only a few other data available regarding the precision (reproducibility) of an interpretation of thyroid FNA demonstrating cyst fluid only to be nondiagnostic. The 2007 Year End Summary Report of the College of American Pathologists' Interlaboratory Comparison Program in Non-Gynecologic Cytopathology does not report results for this interpretative category, but when participants were presented with slides with a reference category of “negative” and interpretation of “goiter,” >5% of pathologists and cytotechnologists classified such negative cases as “cystic lesion, nondiagnostic.”5 Just as others have recommended stringent criteria for FNA sufficiency,6 so too is it recommended that stringent criteria be used for the interpretation of cyst fluid only. The use of this term should be restricted to cases that present with abundant macrophages and/or pigmented macrophages, and inadequate numbers of follicular cells. Possibly, quality control benchmarks could be developed for this interpretive category.7
The amount of colloid, number of macrophages and pigment-laden macrophages, and amount of blood and inflammatory cells were not found to be predictive of a malignant histologic outcome. The sole morphologic feature of malignancy was the presence of atypical features in follicular epithelium, if present. In 4 of 10 cases of ipsilateral carcinoma measuring ≥1 cm, such atypical epithelium could be identified, but it is uncertain whether this epithelium was necessarily derived from the carcinoma. At least some atypical cells from thyroid cysts do not have a malignant origin, but instead originate from cyst-lining cells and resemble classic epithelial repair (Fig. 4).8 Their distinction from cystic papillary thyroid carcinoma is not always possible. The diagnostic phraseology “atypical epithelial cells, cannot exclude papillary thyroid carcinoma” has been proposed for the reporting of rare atypical epithelial cells in a cystic thyroid nodule, with the presence of intranuclear cytoplasmic inclusions, squamoid cytoplasm, and psammoma bodies to be stressed as key morphologic indicators of malignancy.9
In summary, the results of the current study indicate that the only cytomorphologic characteristic predictive of subsequent malignancy in cyst fluid only cases is the presence of follicular epithelium with atypical or suspicious features. Therefore, cases containing atypical epithelial cells should not be categorized as nondiagnostic or cyst fluid only, but rather be diagnosed as atypical or suspicious.
Conflict of Interest Disclosures
Dr. Jaragh was supported by a fellowship grant from the Kuwait Institute for Medical Specialization, Kuwait.
- 1The Papanicolaou Society of Cytopathology Task Force on Standards of Practice. Guidelines of the Papanicolaou Society of Cytopathology for the examination of fine-needle aspiration specimens from thyroid nodules. Diagn Cytopathol. 1996; 15: 84-89.
- 2National Cancer Institute Thyroid Fine Needle Aspiration (FNA) State of the Science Conference. October 22-23, 2008. Available at: http://thyroidfna.cancer.gov/ Accessed December 23, 2008.
- 3Diagnostic terminology and morphologic criteria for cytologic diagnosis of thyroid lesions: a synopsis of the National Cancer Institute thyroid fine-needle aspiration state of the science conference. Diagn Cytopathol. 2008; 36: 425-437., , , et al.
- 5Interlaboratory Comparison Program in Non-Gynecologic Cytopathology (NGC) 2007 Year End Summary Report. Available at: http://www.cap.org/apps/docs/committees/cytopathology/2007_ngc_program_year_end_summary.pdf Accessed December 23, 2008.