| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Article |
Division of Endocrinology (E.K.A., J.H., E.M.), Departments of Medicine, Radiology (C.B.B., P.M.D., M.F.), and Pathology (E.C.), Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts 02115
Address all correspondence and requests for reprints to: Erik K. Alexander, M.D., Thyroid Division, Brigham and Women’s Hospital, H.I.M. Building, 77 Avenue Louis Pasteur, Room 560, Boston, Massachusetts 02115. E-mail: ekalexander{at}partners.org.
Abstract
Thyroid nodules are common. Evaluation of patients with thyroid nodules typically includes fine needle aspiration biopsy (FNA), an approach that has proven to be accurate for the detection of thyroid cancer. Although the majority of biopsies are adequate for a cytological diagnosis, up to 20% will be insufficient or nondiagnostic. Current opinion suggests that such aspirates should be repeated, although no systematic study has investigated the usefulness of this approach, especially when ultrasound guidance is used to direct the initial FNA. We sought to define the predictors and optimal follow-up strategy for initial nondiagnostic ultrasound-guided FNAs of thyroid nodules.
Data were collected for all patients at the Brigham and Women’s Hospital Thyroid Nodule Clinic between 1995–2000 who underwent ultrasound-guided FNA of a thyroid nodule. All patients with nondiagnostic cytology were advised to return for a repeat ultrasound-guided FNA. Patient age, gender, nodule size, cystic content, solitary vs. multinodular thyroid, and nodule location were documented and evaluated as possible predictors of a nondiagnostic biopsy in a multivariable model. The rate of diagnostic cytology obtained on repeat ultrasound-guided FNA was calculated.
A total of 1128 patients with 1458 nodules were biopsied over a 6-yr period. A total of 1269 aspirations (950 patients) were diagnostic, and 189 (178 patients) were nondiagnostic. The cystic content of each nodule was the only significant independent predictor of nondiagnostic cytology (P < 0.001). The fraction of specimens with initial nondiagnostic cytology increased with greater cystic content (P < 0.001 for trend). A diagnostic ultrasound-guided FNA was obtained on the first repeat biopsy in 63% of nodules and was inversely related to increasing cystic content of each nodule (P = 0.03). One hundred and nineteen patients with 127 nodules returned for follow-up as advised, and malignancy was documented in 5%.
Despite ultrasound-guided FNA, there remains a significant risk of initial nondiagnostic cytology, largely predicted by the cystic content of each nodule. Repeat aspiration is often successful and should be the standard approach to such nodules, given their risk of malignancy.
THYROID NODULES ARE present in up to 50% of the adult population, increasing in prevalence with age (1). The evaluation of patients with thyroid nodules greater than or equal to 1 cm in diameter typically includes measurement of serum TSH and fine needle aspiration biopsy (FNA). This approach has proven to be accurate for the detection of thyroid cancer (2). Although the majority of biopsies are adequate for a cytological diagnosis, 5–20% will be insufficient or nondiagnostic (2, 3, 4, 5). Current opinion suggests that nondiagnostic aspirates should be repeated, as such nodules may be malignant (6, 7, 8, 9).
Prior studies have reported that a repeat palpation-guided FNA of nodules with nondiagnostic cytology on initial palpation-guided FNA is diagnostic up to 50% of the time (10). Other studies have documented the benefit of using ultrasound to guide a repeat FNA when the initial palpation-guided FNA is nondiagnostic (4, 11). A growing number of endocrinologists, however, use ultrasound-guided FNA of thyroid nodules as their initial diagnostic procedure, given its excellent ability to identify discrete nodules and direct needle localization (4, 5, 12).
Although the nondiagnostic rate of ultrasound-guided FNA is reported to be lower than the nondiagnostic rate for palpation-guided FNA, approximately 4–16% will continue to be inadequate for cellular diagnosis (4, 5, 12). The algorithm for managing these nondiagnostic cases has not been established. In particular, only small, nonsystematic studies have investigated whether a repeat ultrasound-guided FNA of nodules with nondiagnostic cytology on initial aspiration would show the same degree of success as that noted above (13, 14). We sought to define the prevalence, predictors, and optimal follow-up strategies for initial nondiagnostic ultrasound-guided FNAs of thyroid nodules.
Materials and Methods
We reviewed the records of all patients referred to the triple discipline (endocrine, radiology, and cytology) Thyroid Nodule Clinic at the Brigham and Women’s Hospital for evaluation of nodular thyroid disease between 1995–2000 inclusive. All patients underwent ultrasound examination of the thyroid by a radiologist and ultrasound-guided FNA by an endocrinologist of nodules 1 cm or greater in maximal diameter. All patients with nondiagnostic cytology (<6 groups of cells containing >15 cells each, without evidence of cellular atypia) were advised to return for a repeat ultrasound-guided FNA. The study population included all patients with nondiagnostic cytology on the initial visit who returned for repeat ultrasound-guided FNA during the 6-yr study period.
Thyroid ultrasonography was performed by one of three radiologists (C.B.B., M.C.F., or P.M.D.) using a 5- to 15-MHz transducer. The length, width, and depth of each nodule were reported, and each nodule was classified as solid or less than 25%, 25–50%, 50–75%, or more than 75% cystic. Nodules were further classified as solitary (only one nodule 
1 cm on ultrasound examination) or part of a multinodular thyroid (two or more nodules 1 cm on ultrasound examination), and the location of each nodule was documented (isthmus, left lower, left middle, left upper, right lower, right middle, or right upper). Ultrasound-guided FNA was performed by an endocrinologist using a 25-gauge needle with three or four aspirations per nodule. For cystic nodules, ultrasound and color-flow Doppler were used to direct sampling of the solid portion that showed evidence of tissue perfusion. If cystic fluid was aspirated, it was sent for separate cytological evaluation.
All specimens were processed using the Thin-Prep technique. Each needle was rinsed with CytoLyt solution (Cytyc Corp., Boxborough, MA) into a 50-cc conical tube and centrifuged at 1880 rpm for 10 min. If the cell pellet was bloody, it was resuspended in CytoLyt and recentrifuged. The cell pellet was then transferred to a vial containing PreservCyt solution (Cytyc Corp.), and 2 thin layer slides prepared with the ThinPrep 2000 (Cytyc Corp.). If sufficient cellular material was available, a portion was retained and fixed in 10% formalin, embedded in paraffin, and processed for cell block sections. The ThinPrep slides were stained with a modified Papanicolaou stain, and cell block sections were stained with hematoxylin and eosin. All slides were reviewed by a cytotechnologist and interpreted by a cytopathologist at Brigham and Women’s Hospital. Specimens were considered nondiagnostic if insufficient cellular material (<6 groups of cells containing >15 cells each) was present and no evidence of cellular atypia was found.
The age and gender of each patient as well as the most recent serum TSH concentration were recorded at the time of initial examination. Repeat aspiration was considered diagnostic if specimens were classified as 1) benign (six or more cellular groups without atypical cells), 2) atypical (specimens with atypical cells of undetermined significance, not definitively benign or concerning enough to be suspicious for malignancy), 3) suggestive of follicular neoplasm (a predominance of microfollicles, classically a ringlet of thyroid follicular cells), 4) suggestive of a Hurthle cell neoplasm (specimen composed almost exclusively of Hurthle cells), 5) suspicious for papillary carcinoma (specimens with some, but not all, of the features of papillary carcinoma, including crowded cells, enlarged nuclei with pale, powdery chromatin, nuclear grooves, nuclear pseudoinclusions, distinct nucleoli, papillary structures, psammoma bodies, and "sticky colloid"), or 6) positive for papillary carcinoma (those cases with most of the above features of papillary carcinoma). Patients with atypical cytology where recommended to return for repeat FNA of such nodules in 3–6 months.
Permission from the investigational review board of the Brigham and Women’s Hospital was obtained to perform this investigation.
Statistics
Descriptive statistics are presented according to nodule or according to patient as appropriate and were compared using the two-sample t test (for continuous data) or the 
2 test (for categorical data). The main outcome for the multiple variable predictive model was a nondiagnostic ultrasound-guided FNA, as defined above. Multinomial logistic regression analysis was performed using SPSS software (version 11.0, SPSS, Inc., Chicago, IL). Potential predictors were age, gender, nodule size (millimeters; analyzed as a continuous variable), cystic content (solid or <25%, 25–50%, 50–75%, or >75% cystic), solitary vs. multinodular thyroid, and nodule location. Significance was accepted at P < 0.05.
Results
A total of 1128 patients were examined at the Thyroid Nodule Clinic between 1995 and 2000, and 1458 nodules were biopsied. A total of 1269 aspirations (87%) in 950 patients were diagnostic and were classified into 1 of 6 cytological diagnoses: 929 were benign, 74 were atypical, 80 were suspicious for papillary carcinoma, 47 were positive for papillary carcinoma, 3 were positive for metastatic carcinoma, and 136 were suggestive of follicular or Hurthle cell neoplasm.
There were 189 initial nondiagnostic aspirations (13%) in 178 patients. The baseline demographic and ultrasound characteristics of these 178 patients and their nodules were compared with those of patients with diagnostic cytology (Table 1
). Patients and nodules with initial nondiagnostic cytology were not different from those with diagnostic cytology, except with respect to the cystic content of the nodules. Age, gender, nodule size, nodule location, and solitary vs. multinodular status were not predictive of a nondiagnostic ultrasound-guided FNA. Upon multivariate analysis, the cystic content of each nodule was the only significant, independent predictor of nondiagnostic cytology (P < 0.001). Furthermore, the fraction of specimens with initial nondiagnostic FNA increased with greater cystic content, ranging from 8% in solid nodules to 36% in nodules more than 75% cystic (P < 0.001 for trend; Table 2).
|
|
A diagnostic ultrasound-guided FNA was obtained on the first repeat biopsy in 63% of nodules. The percentage of diagnostic reaspirations decreased as the cystic content increased (P = 0.03; Table 3). Seventy-six percent of initially nondiagnostic solid nodules were diagnostic on repeat biopsy compared with 48% of nodules more than 75% cystic. Thirty-eight nodules were again nondiagnostic on repeat ultrasound-guided FNA. Ten of these underwent a third aspiration, and three were successful (30%).
|
Discussion
Nondiagnostic FNAs of thyroid nodules remain a significant clinical dilemma. Despite improved aspiration techniques and ultrasound guidance, up to 20% of initial aspirations may be nondiagnostic (2, 3, 4, 5). It is clear that such nodules are not implicitly benign (8), and further evaluation is required. Over a 6-yr period, we experienced a 13% rate of nondiagnostic ultrasound-guided FNAs of thyroid nodules. The highest proportion of nondiagnostic specimens occurred in nodules with the greatest cystic content, a finding consistent with prior studies (14, 15). Although one study has reported near-perfect success in obtaining diagnostic specimens using ultrasound guidance in such patients (16), we were not able to match these results despite apparently similar techniques and equipment. Cystic nodules remained a significant diagnostic problem among our cohort, with nondiagnostic specimens reaching 36% in nodules more than 75% cystic.
Other factors have been reported to influence the success of FNA, including small nodule size, position of the nodule within the thyroid, and patient age and body habitus (5, 17). Our data suggest otherwise. We found that cystic content of the nodule was the only independent predictive factor for nondiagnostic specimens (P < 0.001). Nonetheless, even FNAs of solid nodules were nondiagnostic 8% of the time, suggesting that a low rate of nondiagnostic specimens is inherent to the procedure itself, confirming other results (4, 5).
When evaluating an initially nondiagnostic ultrasound-guided FNA, repeat ultrasound-guided FNA was highly successful, providing diagnostic specimens in 63% of cases. As with initial aspirations, the success of repeat aspirations decreased as the cystic content of the nodule increased (P = 0.03). Despite this, however, diagnostic cytology was obtained for nearly 50% of nodules more than 75% cystic, indicating that a repeat ultrasound-guided FNA is warranted for any initial nondiagnostic aspiration. If the specimen remains nondiagnostic, our limited data suggest that a third aspiration is less likely to be diagnostic, but may still prove useful.
Previous studies report the incidence of malignancy to be 5–10% among nondiagnostic thyroid nodules (8, 9). Our study confirms this finding, as 5% of nondiagnostic specimens from our dataset were malignant (all papillary carcinoma). Thus, persistent evaluation of initially nondiagnostic thyroid nodules is mandated and should by pursued promptly.
We acknowledge potential concerns related to our study methods that should be kept in mind. Patients returning for follow-up ultrasound-guided FNA were not selected randomly, allowing for potential physician (or patient) selection bias. However, at the time of their initial nondiagnostic aspiration, all patients were instructed to arrange follow-up examinations regardless of patient or nodule characteristics, and those who complied were no different from those who did not except with regard to nodule size. Second, in determining the malignancy rate among those patients with initial nondiagnostic ultrasound-guided FNA, we acknowledge the potential for false negative cytology reports in those specimens diagnosed as no malignant cells or considered benign because the nodules were smaller on repeat ultrasound. However, a review of 880 benign nodules assessed by ultrasound-guided FNA at our institution over the last 6 yr has shown a less than 1% false negative rate in nodules adequately sampled with ultrasound guidance (unpublished data). Additionally, prior studies confirm a low malignant risk among nodules that are stable or decrease in size over time (7, 18, 19).
In summary, nondiagnostic FNAs remain a significant problem in the evaluation of thyroid nodules. Despite ultrasound guidance, clinicians should advise patients that there is a 5–15% risk of initial nondiagnostic specimens, largely predicted by the cystic content of the nodule. However, repeat ultrasound-guided FNA often provides an adequate specimen and should be the standard approach to such nodules, given their risk of malignancy. Although our series is small, our data suggest that physicians should also consider performing an additional third aspiration on the residual nondiagnostic nodules before other treatment modalities are pursued.
Acknowledgments
We are indebted to Drs. P. Reed Larsen and Robert D. Utiger for helpful review of this manuscript.
Footnotes
This work was supported by NIH Training Grant DK-07529 and a research grant from the Endocrine Fellows Foundation.
Abbreviation: FNA, Fine needle aspiration biopsy.
Received June 4, 2002.
Accepted August 8, 2002.
References
This article has been cited by other articles:
 |
|
 |
|
  E. B. Strauss, A. Iovino, and S. Upender Simultaneous Fine-Needle Aspiration and Core Biopsy of Thyroid Nodules and Other Superficial Head and Neck Masses Using Sonographic Guidance Am. J. Roentgenol., June 1, 2008; 190(6): 1697 - 1699. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C Cappelli, I Pirola, M Castellano, E Gandossi, E De Martino, A Delbarba, B Agosti, A Tironi, and E A. Rosei Fine needle cytology of complex thyroid nodules Eur. J. Endocrinol., October 1, 2007; 157(4): 529 - 532. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. C. Frates, C. B. Benson, P. M. Doubilet, E. Kunreuther, M. Contreras, E. S. Cibas, J. Orcutt, F. D. Moore Jr., P. R. Larsen, E. Marqusee, et al. Prevalence and Distribution of Carcinoma in Patients with Solitary and Multiple Thyroid Nodules on Sonography J. Clin. Endocrinol. Metab., September 1, 2006; 91(9): 3411 - 3417. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. C. Frates, C. B. Benson, J. W. Charboneau, E. S. Cibas, O. H. Clark, B. G. Coleman, J. J. Cronan, P. M. Doubilet, D. B. Evans, J. R. Goellner, et al. Management of Thyroid Nodules Detected at US: Society of Radiologists in Ultrasound Consensus Conference Statement Radiology, December 1, 2005; 237(3): 794 - 800. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Dossing, F. N. Bennedbaek, and L. Hegedus Effect of ultrasound-guided interstitial laser photocoagulation on benign solitary solid cold thyroid nodules - a randomised study Eur. J. Endocrinol., March 1, 2005; 152(3): 341 - 345. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. J. Mandel A 64-Year-Old Woman With a Thyroid Nodule JAMA, December 1, 2004; 292(21): 2632 - 2642. [Full Text] [PDF]
|
|
|
|
|
|
F. N. Bennedbaek and L. Hegedus Treatment of Recurrent Thyroid Cysts with Ethanol: A Randomized Double-Blind Controlled Trial J. Clin. Endocrinol. Metab., December 1, 2003; 88(12): 5773 - 5777. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. L. Titton, D. A. Gervais, G. W. Boland, M. M. Maher, and P. R. Mueller Sonography and Sonographically Guided Fine-Needle Aspiration Biopsy of the Thyroid Gland: Indications and Techniques, Pearls and Pitfalls Am. J. Roentgenol., July 1, 2003; 181(1): 267 - 271. [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 |
