| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Departments of Nuclear Medicine (L.L., L.D.P.F., D.L.G., T.D., C.G., A.A.), General and Gastrointestinal Surgery (F.M.), Pathology (C.H.), and Endocrinology (G.T.) and the Quantitative Medical Imaging Research Unit, INSERM U-494 (L.L., G.H., B.F., A.A.), Pitié Hospital, 75013 Paris; the Department of Pathology (B.F.), Ambroise Paré Hospital, 92100 Boulogne-Billancourt, France
Address all correspondence and requests for reprints to: Dr. L. Leenhardt, Service Central de Médecine Nucléaire, Groupe Hospitalier Pitié-Salpêtrière, 83 boulevard de l’Hôpital, 75013 Paris, France.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
From December 1989 to November 1995, 1741 consecutive patients
with thyroid nodules were referred for FNAB in our department. Among
them, 450 (26%) patients presented with nonpalpable nodules and
underwent US-FNAB. These nodules had been diagnosed by US performed for
various reasons (Table 1
). Preliminary
clinical examination was performed by 2 experienced endocrinologists
(L.L. and T.D.) before FNAB to confirm that the nodules were not
palpable (387 cases; 86%) or too deeply located (63 cases; 14%) to be
reliably biopsied under palpation-guided FNAB. The series included 372
females and 78 males, with a mean age of 49.5 yr (ranging from 16–83
yr).
|
US of the thyroid gland was performed using a real-time ultrasonographic scanner (Advanced Technology Laboratory, Washington DC) with a 7.5-MHz linear transducer. All scans were performed in our department by an experienced echographist (T.D.). Among the 450 nodules included in the series, 136 (30%) were solitary, whereas 314 (70%) were associated with other nodules. The echo structure (solid, mixed, or cystic), echogenecity (hyperechoic, isoechoic, or hypoechoic), calcification (presence or absence), and margin (well defined or blurred) were assessed. Cystic nodules were defined as anechoic nodules. The nodule mean diameter was considered the arithmetic mean of the measured length, width, and depth. Volume was estimated through the relation: V = (length x width x depth)/2. After a logarithmic transformation, the nodule volume distribution was Gaussian (mean = -0.058; SD = 0.46). According to the nodule volume, the series was partitioned into 10 quantiles for statistical analysis.
US-FNAB were all performed by the same physician (L.L.), using a 25-gauge needle with a free hand technique. For a partially cystic nodule, the biopsy sampling was directed to the solid portion of the nodule. In patients with more than one nodule, FNAB on only the largest one was retained. During the course of the study, some patients were subjected to repetitive FNAB. In such cases, only the result of the last FNAB was considered.
Material was smeared on slides and stained by May-Grunwald-Giemsa stain. The cytological analysis was always performed by the same pathologist (B.F.). Insufficient cytological material (ICM) was defined as the presence of less than six follicular thyroid cell clusters on the slides. Conversely, cytological material was defined as adequate when six or more thyroid cell clusters were obtained. Adequate cytological material (ACM) was classified as benign, malignant, or suspicious. Benign cytological results corresponded to colloid or macrofollicular adenomas, nodular and/or cystic goiters, or thyroiditis. Malignant cytological results corresponded to diagnosis of papillary carcinoma, high grade follicular carcinoma, or medullary or anaplastic carcinoma. Suspicious cytological results corresponded to smears with a high cellularity, microacinar formation, a scanty or absent colloid, large nuclei and noticeable anisocaryosis, and/or a dominance of Hurthle cells. Suspicious cytological results were handled as malignant diagnoses for statistical analysis.
Among the 450 patients, 94 (21%) underwent surgery. Indications for surgery were FNAB diagnosis of malignant or suspicious lesions in 40 of 94 cases (43%), supracentimetric or isolated cold nodule in 24 cases (26%), simultaneous presence of a palpable nodule in a multinodular gland in 16 cases (17%), and miscellaneous reasons in 14 cases (15%; 2 hyperthyroidism, 4 cervical adenopathy, and 7 increasing size of nodule and suspicious echographic features). The final histological diagnosis (according to the WHO classification) after surgical removal was considered the gold standard (9). Seven occult histological carcinomas found outside the biopsied nodule were not taken into account and were considered incidental.
Statistical analyses were performed using the SPSS (SPSS, Inc., Chicago, IL) software package. All statistical analyses involving nodule volume were performed on logarithm-transformed volume data. The relationships between the proportion of ACM and various ultrasonographic features, including nodule size, were explored using logistic regression. A logistic model including such ultrasonographic features to predict histological diagnosis was also designed.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
). Table 2
indicates that 42 of 293 (14%) patients with benign cytological
results were subjected to surgery. The corresponding data for patients
with suspicious, malignant, and ICM results were 24 of 50 (48%), 16 of
22 (73%), and 12 of 85 (14%), respectively. Among the 94 operated
nodules, 20 (21%) carcinomas were found at final histology,
representing 4% of the population biopsied.
|
Among the 450 nodules subjected to FNAB, 365 (81%) yielded ACM,
and 85 (19%) yielded ICM for diagnosis. The mean diameter of the 450
nodules as measured by ultrasonography ranged from 0.43–3.67 cm (mean,
1.33 cm; median, 1.22), and the mean diameter of the 94 operated
nodules ranged from 0.57–2.6 cm (mean, 1.29 cm; median, 1.23). Table 3 shows the proportion of ACM according
to nodule volume. Considering the partition of the series in ten
quantiles, this proportion varied from 64–91%. The observed variation
did not significantly differ from linearity (P <
0.92), and logistic regression indicated that the proportion of ACM
significantly increased (P < 0.0001) with the nodule
volume.
|
details the distribution of US
parameters in the series. The logistic model indicates that the
proportion of ACM did not significantly differ according to the
echostructure (P < 0.78), the echogenecity
(P < 0.68), or the presence of calcifications
(P < 0.46), but was significantly associated with well
defined margins (P < 0.02).
|
The 20 carcinomas diagnosed at final histology corresponded to 16 papillary, 3 follicular, and 1 medullary carcinomas. The ultrasonographic mean diameter of the 20 carcinomas ranged from 0.63–2.27 cm (mean, 1.23 cm; median, 1.08), and 12 of them (60%) were centimetric or supracentimetric.
Diagnosis of malignancy was successfully made by US-FNAB in 16 cases,
unequivocally made in 9 cases, strongly suggested in 7 cases, and
missed in 4 cases, 3 because of ICM and 1 misdiagnosed as a benign
cytological result (Table 2
). The mean diameters of 2 of 3 missed
malignant nodules because of ICM, were supracentimetric (1.1 and 1.6
cm). The sensitivity of US-FNAB diagnosis in the operated nodule sample
was 94%, and the specificity was 63% (Table 5). As shown in Table 4,
none of the 20 malignant nodules was entirely cystic; 17 (85%) were
homogeneously solid, and 3 were mixed (with cystic and solid
components). Eighteen (90%) carcinomas were hypoechoic, and only 1 was
hyperechoic; 16 (80%) were both homogeneously solid and hypoechoic.
Calcifications and blurred margins were found, respectively, in 10 and
11 of the 20 malignant nodules; 7 malignant and 11 benign nodules
presented both calcifications and blurred margins.
|
). It should be
noted that 45 of the 115 infracentimetric nodules were both solid and
hypoechoic (Table 6). Among these, 16 were operated and 7 corresponded
to carcinomas at final histology. Conversely, only 1 malignant nodule
was found at final histology of the 77 cystic or hyperechoic nodules.
Considering other US parameters, neither nodule size (P
< 0.6), the presence of calcification (P < 0.48), nor
blurred margins (P < 0.25) were significantly
associated with the histological diagnosis. Benign and malignant
nodules had comparable size distributions.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
). In our series, the proportion
of ACM was 85% in centimetric and supracentimetric nodules and 69% in
infracentimetric nodules. US-FNAB allowed the discovery of 10 (50%)
supracentimetric of 20 histological carcinomas (Table 2). Therefore,
our results indicate that supracentimetric nonpalpable nodules benefit
from US-FNAB. Large malignant nodules have been reported to be missed
by palpation (15). In a series of 145 nodules, Witterick et
al. studied the size of 20 malignant nodules confirmed at
histology that escaped detection by palpation. The greatest size of
malignant nodules was 2.1 cm (15). Rosen et al. report 11 of
15 nonpalpable carcinomas with a diameter greater than 1 cm (14). In
our study, 12 nonpalpable malignant nodules were centimetric or
supracentimetric (Table 2). The existence of these nodules, detected by
US exploration, suggests that a simple follow-up neck palpation, as
previously proposed (5, 16), may not be the safest management
strategy.
Our results indicate that besides size, other ultrasonographic
parameters should be considered. In particular, the proportion of ACM
was significantly greater for nodules with well defined margins than
for nodules with blurred margins (Table 4). The echostructure, the
presence of calcifications, and the echogenecity features were not
significantly associated with ACM. The useful contribution of US
guidance that directs the needle toward the solid component in cystic
nodules and avoids a puncture across dense calcifications may explain
our results (7, 13). Thus, we conclude that in this series, centimetric
and supracentimetric size and well defined margins constitute
propitious conditions for the success of the US-FNAB procedure in
nonpalpable nodules.
Considering the high prevalence of US thyroid nodules (4), we believe that a systematic US-FNAB would lead to an unjustified burden for health care. Indeed, there is a low chance of detecting cancers (4% of the biopsied population in the present series), and the prognosis of the papillary thyroid microcarcinoma is good (17). The evaluation of strategies for an optimal management of thyroid nonpalpable nodules is still a matter of debate, and our study, as well as others, has some limitations.
First, the techniques are only evaluated on operated patients.
Such a sample does not reflect the total biopsied population. Moreover,
in most of the series, only a proportion of the malignant and
suspicious cytological results are compared to histological results,
whereas some patients with benign cytology are subjected to surgery. In
our series, the surgical and histological data of 6 patients with
malignant cytological results were unknown despite repeated efforts for
obtaining information. Conversely, 42 patients with benign cytological
results and 12 patients with ICM had surgery for indications detailed
in Subjects and Methods, and 4 of these 54 patients had
carcinoma at final histology (Table 2). The indications for surgery in
these 4 cases were as follows: hyperthyroidism, goiter, suspicious
US-FNAB results of nonpalpable adenopathy detected by ultrasound, and
suspicious cytological results on an associated palpable nodule.
Second, discrepancies in the reported FNAB performances depend on the various ways the suspicious cases are handled in the calculations (18). Some authors only take into consideration benign and malignant cytological results. Others consider suspicious and/or follicular initial diagnosis as a diagnosis of malignancy and classify such cases as true positive even if they correspond to benign microfollicular adenomas (1, 2). In our study, true positive results corresponded to FNAB malignant or suspicious nodules confirmed as carcinomas at final histology.
Third, the value of US-FNAB in nonpalpable nodules has to be balanced
with the clinical context. Associated thyroid disease or symptoms were
useful and led to an appropriate assessment by US followed by US-FNAB
in 12 carcinomas whereas 8 carcinomas were found incidentally, by
screening or for questionable indications of US exploration (Table 1).
Nevertheless, our results led us to study the impact of four US-FNAB
based strategies in the management of nonpalpable thyroid nodules. The
four strategies are shown in Table 6 and
are derived from surgical exploration of suspicious and malignant
cytological results:
The first strategy, US-FNAB performed on all nodules, will be used as a
reference (Table 6). US-FNAB allowed the detection of 16 histological
carcinomas and corresponded to a sensitivity of 94%. In other series
reported in the literature, the sensitivity and specificity of US-FNAB
amounted to 79% and 85%, respectively (6). When reviewing the results
of 18,183 FNAB in 7 large series, Gharib and Goellner indicate a
sensitivity of the method varying from 65–98% and a specificity
varying from 72–100%; the overall accuracy was 95% (1). However, the
studies only concern direct FNAB on palpable nodules. Some authors did
not detect any significant difference between direct and US-guided FNAB
(7). Others indicate that US guidance provides a more precise and
adequate sampling with a lower rate of false negatives (19, 20).
However, the palpable and nonpalpable nodules are hardly comparable in
these studies. Due to the high prevalence of US thyroid nodules, a
systematic US-FNAB performed on all nonpalpable nodules is not
advisable.
The second strategy refers to a systematic US-FNAB performed only on centimetric and supracentimetric nodules. According to this strategy, only 10 carcinomas would have been detected; 8 of the 10 missed carcinomas would have been infracentimetric. Therefore, such a strategy does not appear advisable.
The third strategy would suggest a US-FNAB on solid and hypoechoic nodules. However, the relevance of US features for predicting malignancy is still not clearly defined. Although the low specificity of US screening of thyroid nodules has been underlined by some authors (21, 22, 23), it is generally accepted that solid hypoechoic nodules are suspicious US features (24). In our series, only 31% of the patients would be subjected to FNAB; 13 carcinomas would have been detected, 7 of these being infracentimetric. Conversely, the indication of US-FNAB in cystic or hyperechoic nonpalpable nodules is questionable.
The fourth strategy is to perform US-FNAB on centimetric, supracentimetric, or solid hypoechoic nodules. The interest of such a strategy must be balanced with those of the previous one. With the fourth strategy, 84% of the patients would be subjected to FNAB, but 15 carcinomas would be detected. In our view, this strategy, allowing the detection of carcinomas with the best reliability while avoiding a systematic US-FNAB on all nodules, should be recommended.
In conclusion, US-FNAB must be favorably considered for nonpalpable thyroid nodules, especially for the exploration of centimetric, supracentimetric, or solid hypoechoic nodules. The relative failure of US-FNAB on infracentimetric nodules is outlined and must be balanced with the high proportion of cancers detected in solid and hypoechoic nodules.
|
Acknowledgments |
|---|
Received March 20, 1998.
Revised September 30, 1998.
Accepted October 14, 1998.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  Y. Erbil, U. Barbaros, A. Salmaslioglu, O. Mete, H. Issever, S. Ozarmagan, D. Yilmazbayhan, and S. Tezelman Effect of Thyroid Gland Volume in Preoperative Detection of Suspected Malignant Thyroid Nodules in a Multinodular Goiter Arch Surg, June 1, 2008; 143(6): 558 - 563. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Yeung and J. W. Serpell Management of the Solitary Thyroid Nodule Oncologist, February 1, 2008; 13(2): 105 - 112. [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]
|
|
|
|
 |
|
 C. S. Hollenbeak, I. Lendel, K. S. Beus, J. M. Ruda, and B. C. Stack Jr The Cost of Screening for Synchronous Thyroid Disease in Patients Presenting With Primary Hyperparathyroidism Arch Otolaryngol Head Neck Surg, October 1, 2007; 133(10): 1013 - 1021. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R Jones, R Spendiff, S Fareedi, and P S Richards The role of ultrasound in the management of nodular thyroid disease Imaging, March 1, 2007; 19(1): 28 - 38. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Cappelli, M. Castellano, I. Pirola, D. Cumetti, B. Agosti, E. Gandossi, and E. Agabiti Rosei The predictive value of ultrasound findings in the management of thyroid nodules QJM, January 1, 2007; 100(1): 29 - 35. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Cappelli, M. Castellano, I. Pirola, E. Gandossi, E. De Martino, D. Cumetti, B. Agosti, and E. A. Rosei Thyroid nodule shape suggests malignancy. Eur. J. Endocrinol., July 1, 2006; 155(1): 27 - 31. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M Niedziela Pathogenesis, diagnosis and management of thyroid nodules in children. Endocr. Relat. Cancer, June 1, 2006; 13(2): 427 - 453. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. R. Steele, M. J. Martin, P. S. Mullenix, K. S. Azarow, and C. A. Andersen The Significance of Incidental Thyroid Abnormalities Identified During Carotid Duplex Ultrasonography Arch Surg, October 1, 2005; 140(10): 981 - 985. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Lyshchik, V. Drozd, Y. Demidchik, and C. Reiners Diagnosis of Thyroid Cancer in Children: Value of Gray-Scale and Power Doppler US Radiology, May 1, 2005; 235(2): 604 - 613. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. R. Wienke, W. K. Chong, J. R. Fielding, K. H. Zou, and C. A. Mittelstaedt Sonographic Features of Benign Thyroid Nodules: Interobserver Reliability and Overlap With Malignancy J. Ultrasound Med., October 1, 2003; 22(10): 1027 - 1031. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Hegedus, S. J. Bonnema, and F. N. Bennedbaek Management of Simple Nodular Goiter: Current Status and Future Perspectives Endocr. Rev., February 1, 2003; 24(1): 102 - 132. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. K. Alexander, J. P. Heering, C. B. Benson, M. C. Frates, P. M. Doubilet, E. S. Cibas, and E. Marqusee Assessment of Nondiagnostic Ultrasound-Guided Fine Needle Aspirations of Thyroid Nodules J. Clin. Endocrinol. Metab., November 1, 2002; 87(11): 4924 - 4927. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. S. Ross Nonpalpable Thyroid Nodules--Managing an Epidemic J. Clin. Endocrinol. Metab., May 1, 2002; 87(5): 1938 - 1940. [Full Text] [PDF]
|
|
|
|
|
|
E. Papini, R. Guglielmi, A. Bianchini, A. Crescenzi, S. Taccogna, F. Nardi, C. Panunzi, R. Rinaldi, V. Toscano, and C. M. Pacella Risk of Malignancy in Nonpalpable Thyroid Nodules: Predictive Value of Ultrasound and Color-Doppler Features J. Clin. Endocrinol. Metab., May 1, 2002; 87(5): 1941 - 1946. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Corrias, S. Einaudi, E. Chiorboli, G. Weber, A. Crino, M. Andreo, G. Cesaretti, L. de Sanctis, M. F. Messina, M. Segni, et al. Accuracy of Fine Needle Aspiration Biopsy of Thyroid Nodules in Detecting Malignancy in Childhood: Comparison with Conventional Clinical, Laboratory, and Imaging Approaches J. Clin. Endocrinol. Metab., October 1, 2001; 86(10): 4644 - 4648. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. H. Samuels Editorial: Evaluation and Treatment of Sporadic Nontoxic Goiter--Some Answers and More Questions J. Clin. Endocrinol. Metab., March 1, 2001; 86(3): 994 - 997. [Full Text]
|
|
|
|
 |
|
 E. Marqusee, C. B. Benson, M. C. Frates, P. M. Doubilet, P. R. Larsen, E. S. Cibas, and S. J. Mandel Usefulness of Ultrasonography in the Management of Nodular Thyroid Disease Ann Intern Med, November 7, 2000; 133(9): 696 - 700. [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 |
1 cm), and
histological diagnosis