- Email: [email protected]
Cytologic determinants of well-differentiated thyroid cancer
Cytologic
Determinants Thyroid Melita A. Charles,
MD, Keith S. Heller, MD, New Hyde Park, New York
OBJECTIVE: Fine-needle aspiration biopsy (FNAB) is the preferred diagnostic study for evaluating thyroid nodules. Despite its accuracy, many patients undergo thyroidectomy for benign nodules. This study was undertaken to identify risk factors that might increase the specificity of FNAB. METHODS: Medical records of 422 patients who underwent thyroid surgery between 1966 and 1996 were reviewed. All patients had FNAB prior to surgery. RESULTS: Of the 422 patients, 36% had benign cytology, 46% had indeterminate cytology, and 13% had cancer. In the indeterminate group, 29% of patients had cancer at surgery. Of patients with papillary cytology, 64% had malignancies. Five percent of FNABs were nondiagnostic. Neither age, gender, nor tumor size was associated with increased specificity of FNAB. CONCLUSION: There is no subpopulation of patients with indeterminate FNAB cytology at increased risk of having well-differentiated thyroid cancer. Am J Surg. 1997;174:545-547.0 1997 by Excerpta Medica, Inc.
F
me-needle aspiration biopsy (FNAB) is the diagnostic modality of choice in the evaluation of thyroid nodules.’ It is quite accurate when obviously benign or malignant cells are identified.’ A large subgroup of patients, however, have FNABs that are read as suspicious or indeterminate.2-4 In this group, only 30% will have a malignancy found at surgery.5 In addition, because frozen section is no more accurate than FNAB in detecting thyroid cancer,‘j second surgical procedures may be required if total thyroidectomy is indicated based on the results of the initial thyroid lobectomy. This study is an attempt to identify histologic and clinical characteristics that might improve the accuracy of FNAB.
METHODS A retrospective review of the medical records of all patients who underwent thyroid surgery between 1986 and 1996 was conducted. These patients were treated by two surgeons in a single practice. Only those patients who had FNAB prior to surgery were included in this study. The FNAB reports were retrospectively assigned to one of four From the Head and Neck Service, Department of Surgery, Long Island Jewish Medical Center, New Hyde Park, New York. Requests for reprints should be addressed to Keith S. Heller MD, 200 Middleneck Road, Great Neck, New York 11021. Presented at the 43rd Annual Meeting of the Society of Head and Neck Surgeons, Cancdn, Mexico, April 10-12, 1997.
I’
71 0 1997 by Excerpta All rights reserved.
Medica,
Inc.
of We&Differentiated Cancer
categories (1 = benign, 2 = low suspicion, 3 = suspicious, 4 = malignant) without knowledge of the surgical findings. If more than one FNAB was performed, the more definitive one was used for analysis. Data from 452 patients were collected. Twenty-two patients with clinically obvious metastatic thyroid cancer and 8 patients with cancers other than papillary, follicular, or Hurthle cell were excluded from further analysis. The remaining 422 patients are the subject of this report. Statistical significance was determined using Fisher’s exact test.
RESULTS Of the 422 patients, 358 (85%) were female with a median age of 47 (range 14 to 84); and 64 (15%) patients were male with a median age of 49 (range 18 to 82). In all, 107 (30%) women and 26 (41%) men had cancer. The results of the FNABs and corresponding surgical findings are summarized in Table 1. Twenty-two (5%) FNABs were nondiagnostic, 150 (36%) FNABs were class 1, 43 (10%) were class 2, 153 (36%) were class 3, and 54 (13%) were class 4. Of patients in class 1, 7% had cancer as did 23% of class 2, 37% of class 3, and 94% of class 4. The difference in the incidence of cancer among the four different classes of FNABs is highly significant (I’
COMMENTS Nodular thyroid disease affects approximately 4% to 7% of the US population.7 However, thyroid cancer occurs in fewer than 5% of nodules.’ Although FNAB of the thyroid was first described by Martin and Ellis in the 1930~,~ it was not widely used until the studies of Hamburger, Rosen, LoGerfo, and others proved its usefulness.‘-12 The use of FNAB has led to a 25% decrease in the number of patients undergoing thyroidectomy and a 15% to 30% increase in the incidence of cancer in the nodules of patients undergoing surgery. I1312For these reasons FNAB has become the 0002-961 O/97/$1 7.00 PII SOOO2-961 0(97)00170-O
545
CYTOLOGIC
DETERMINANTS
TABLE
OF THYROID
CANCEFUCHARLESAND-
I Surgical
Pathology
FNAB
Papillary Cancer
Class 1 Class 2 Class 3 Class 4 Nondiagnostic Total
6 5 30 45 3 89
TABLE
on Fine Needle
Papillary Follicular
(4%) (12%) (20%) (83%) (14%) (21%)
4 1 14 5 2 26
Cell Type
Follicular Papillary Hikthle Not specified Total
(3%) (2%) (9%) (9%) (9%) (6%)
1 2 6 0 0 9
on Fine Needle
Aspiration
Class
I
2
Class
95 (7%) 0 11 (0%) 44 (9%) 150
(FNAB) Hiirthle Cancer
(1%) (5%) (4%)
0 2 6 1 0 9
(2%)
of patients
JOURNAL
OF SURGERY”
VOLUME
174
Biopsy Class
33 (15%) 0 9 (44%) 1 (100%) 43
diagnostic modality of choice in the management of thyroid nodules. Fine needle aspiration biopsy is quite accurate in distinguishing cancer from benign disease. Reported false-negative rates range from 1% to 11% and false-positive rates from 1% to 8%.’ In this study, we report a false-negative rate of 7% and a false-positive rate of 6% which is well within these ranges. The number of nondiagnostic aspirates in this series was 5%, which compares favorably with the reported rates of 2% to 21%.’ This low incidence may be due to the fact that repeat aspirates were frequently performed if the initial FNAB was nondiagnostic. Giuffrida and Gharib13 reported that 64% to 73% of aspirates were benign, 3% to 4% malignant, and 11% to 17% suspicious or indeterminate. In our series, 36% of FNABs are benign, 13% malignant, and 46% indeterminate or suspicious. The discrepancy in these figures when compared with other reported series arises from the fact that most patients in this report were referred after evaluation by their primary physician or endocrinologist suggested the need for surgery. Many had positive FNABs at the time of referral. Thirty-four percent of patients with an indeterminate biopsy (class 2 or 3) were found to have cancer at surgery compared with the reported average of 30%.4,5 Biopsies with cells showing papillary differentiation but not meeting strict criteria for malignancy were associated with a diagnosis of cancer in 56% of patients. For that reason, we recommend that all patients whose FNABs show papillary features undergo surgery. In the remaining patients with indeterminate nonpapillary cytology, the incidence of malignancy was 29%. We attempted to study other factors that might be associated with an increased risk of malignancy in this subgroup of patients. Neither age, gender, cell type, nor tumor size correlated with an increased incidence of malignancy. Because the majority of patients in the indeterminate group who undergo surgery are found to have benign disTHE AMERICAN
Biopsy
Follicular Cancer
Type
Numbers in parentheses indicate percentage Twenty-two FNABs were nondiagnostic.
546
Aspiration
Cancer Variant
Benign 139 33 97 3 17 289
(5%) (4%) (2%) (2%)
(93%) (77%) (63%) (6%) (77%) (68%)
II Cell
FNAB
Based
57 27 52 17 153
(FNAB) 3
(35%) (56%) (23%) (41%)
Class 0 54 (94%) 0 0 54
4
Total 185 81 72 62 400
with mct?r,
ease, other authors have tried to identify factors that could further subdivide this group and predict the likelihood of thyroid cancer. Tyler et all4 discussed the management of patients with indeterminate FNABs. In their study, 104 patients with indeterminate cytology were categorized into four groups based on clinical suspicion of cancer; 81 patients underwent thyroidectomy. The majority of invasive cancers were found in patients whose lesions had been classified as suspicious for papillary cancer. In addition, patients over 50 years of age who had a cytologic diagnosis of follicular or Hiirthle cell neoplasm were also identified as a subgroup with increased risk for cancer. Hamming et al’ evaluated the accuracy of preoperative FNAB by dividing patients into high, moderate, and low categories of suspicion on clinical grounds without previous knowledge of either the histologic or cytologic result. Clinical factors including rapid growth, nodule size, age, and radiation history were used to assign patients to each group. The rate of malignancy in the high suspicion group was 71% whereas the rate in the moderate and low suspicion groups were 14% and ll%, respectively. The presence of follicular cells in an FNAB may indicate a neoplasm in 10% to 60% of patients4 In this study, 17% of patients with follicular cells in the FNAB were shown to have cancer. Similarly, only 22% of patients with Htirthle cells in this series had cancer. This suggests that the finding of follicular cells or Hiirthle cells should not necessarily be considered an indication for surgery. Furthermore, most patients with follicular cells on FNAB who have cancer have papillary rather than follicular cancer. Based on cytologic criteria alone, we attempted to separate patients whose aspirates were suspicious (class 3) from those that were only slightly suspicious (class 2). The difference in the incidence of cancer in these two groups is not significant, suggesting that based on current cytologic criteria it is not possible to stratify the degree of suspicion once those patients with papillary cytology are excluded. NOVEMBER
1997
1 CYTOLOGIC
REFERENCES 1. Caruso D, Mazzaferri EL. Fine needle aspiration biopsy in the management of thyroid nodules. Endocrinologist. 1991;1:194202. 2. Hamming JF, Goslings BM, vanSteenis 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. Arch Intern Med. 1990;150:113-116. 3. Gharib H. Fine needle aspiration biopsy of thyroid nodules: advantages, limitations and effect. Mayo Clin Proc. 1994;69:44-49. 4. DeJong SA, Demeter JG, Castelli M, et al. Follicular cell predominance in the cytologic examination of dominant thyroid nodules indicates a sixty percent incidence of neoplasia. Surgery. 1990;108:794-799. 5. Gharib H, Goellner JR, Johnson DA. Fine needle aspiration cy tology of the thyroid. Clin Lab Med. 1993;13:699-709. 6. Mazzaferri EL. Management of a solitary thyroid nodule. NEJM. 1993;328:189-195. 7. Gharib H, Goellner JR. Fine needle aspiration biopsy of the thy raid: an appraisal. Ann Intern Med. 1993;118:282-289.
THE
AMERICAN
DETERMINANTS
OF THYROID
CAiCERKHARLES
AND
HELLER~:
8. Martin HE, Ellis EB. Biopsy by needle puncture and aspiration. Ann Surg. 1930;92:169-181. 9. Miller TM, Hamburger ]I, Kim SR. The impact of needle biopsy on the preoperative diagnosis of thyroid nodules. Henry Fcxd Hasp MedJ. 1980;28:145-148. 10. Rosen IB, Wallace C, Strawridge HG, et al. Reevaluation of needle aspiration cytology in detection of thyroid cancer. Surgery. 1981;90:747-756. 11. Colacchio TA, LoGerfo P, Feind LR. Fine needle cytologic diagnosis of thyroid nodules. Review and report of 300 cases. Am J Surg. 1980;149:568-571. 12. Hamberger B, Gharib H, Melton LJ III, et al. Fine needle aspiration biopsy of thyroid nodules. Impact on thyroid practice and cost of care. Am J Med. 1982;73:381-384. 13. Giuffrida D, Gharib H. Controversies in the management of cold, hot, and occult thyroid nodules. Am J Med. 1995;99:642-650. 14. Tyler DS, Winchester DJ, Caraway NP, et al. Indeterminate fine-needle aspiration biopsy of the thyroid: identification of subgroups at high risk for invasive carcinoma. Surgery. 1994;16:10541060.
JOURNAL
OF
SURGERY@
VOLUME
174
NOVEMBER
1997
547
Determinants Thyroid Melita A. Charles,
MD, Keith S. Heller, MD, New Hyde Park, New York
OBJECTIVE: Fine-needle aspiration biopsy (FNAB) is the preferred diagnostic study for evaluating thyroid nodules. Despite its accuracy, many patients undergo thyroidectomy for benign nodules. This study was undertaken to identify risk factors that might increase the specificity of FNAB. METHODS: Medical records of 422 patients who underwent thyroid surgery between 1966 and 1996 were reviewed. All patients had FNAB prior to surgery. RESULTS: Of the 422 patients, 36% had benign cytology, 46% had indeterminate cytology, and 13% had cancer. In the indeterminate group, 29% of patients had cancer at surgery. Of patients with papillary cytology, 64% had malignancies. Five percent of FNABs were nondiagnostic. Neither age, gender, nor tumor size was associated with increased specificity of FNAB. CONCLUSION: There is no subpopulation of patients with indeterminate FNAB cytology at increased risk of having well-differentiated thyroid cancer. Am J Surg. 1997;174:545-547.0 1997 by Excerpta Medica, Inc.
F
me-needle aspiration biopsy (FNAB) is the diagnostic modality of choice in the evaluation of thyroid nodules.’ It is quite accurate when obviously benign or malignant cells are identified.’ A large subgroup of patients, however, have FNABs that are read as suspicious or indeterminate.2-4 In this group, only 30% will have a malignancy found at surgery.5 In addition, because frozen section is no more accurate than FNAB in detecting thyroid cancer,‘j second surgical procedures may be required if total thyroidectomy is indicated based on the results of the initial thyroid lobectomy. This study is an attempt to identify histologic and clinical characteristics that might improve the accuracy of FNAB.
METHODS A retrospective review of the medical records of all patients who underwent thyroid surgery between 1986 and 1996 was conducted. These patients were treated by two surgeons in a single practice. Only those patients who had FNAB prior to surgery were included in this study. The FNAB reports were retrospectively assigned to one of four From the Head and Neck Service, Department of Surgery, Long Island Jewish Medical Center, New Hyde Park, New York. Requests for reprints should be addressed to Keith S. Heller MD, 200 Middleneck Road, Great Neck, New York 11021. Presented at the 43rd Annual Meeting of the Society of Head and Neck Surgeons, Cancdn, Mexico, April 10-12, 1997.
I’
71 0 1997 by Excerpta All rights reserved.
Medica,
Inc.
of We&Differentiated Cancer
categories (1 = benign, 2 = low suspicion, 3 = suspicious, 4 = malignant) without knowledge of the surgical findings. If more than one FNAB was performed, the more definitive one was used for analysis. Data from 452 patients were collected. Twenty-two patients with clinically obvious metastatic thyroid cancer and 8 patients with cancers other than papillary, follicular, or Hurthle cell were excluded from further analysis. The remaining 422 patients are the subject of this report. Statistical significance was determined using Fisher’s exact test.
RESULTS Of the 422 patients, 358 (85%) were female with a median age of 47 (range 14 to 84); and 64 (15%) patients were male with a median age of 49 (range 18 to 82). In all, 107 (30%) women and 26 (41%) men had cancer. The results of the FNABs and corresponding surgical findings are summarized in Table 1. Twenty-two (5%) FNABs were nondiagnostic, 150 (36%) FNABs were class 1, 43 (10%) were class 2, 153 (36%) were class 3, and 54 (13%) were class 4. Of patients in class 1, 7% had cancer as did 23% of class 2, 37% of class 3, and 94% of class 4. The difference in the incidence of cancer among the four different classes of FNABs is highly significant (I’
COMMENTS Nodular thyroid disease affects approximately 4% to 7% of the US population.7 However, thyroid cancer occurs in fewer than 5% of nodules.’ Although FNAB of the thyroid was first described by Martin and Ellis in the 1930~,~ it was not widely used until the studies of Hamburger, Rosen, LoGerfo, and others proved its usefulness.‘-12 The use of FNAB has led to a 25% decrease in the number of patients undergoing thyroidectomy and a 15% to 30% increase in the incidence of cancer in the nodules of patients undergoing surgery. I1312For these reasons FNAB has become the 0002-961 O/97/$1 7.00 PII SOOO2-961 0(97)00170-O
545
CYTOLOGIC
DETERMINANTS
TABLE
OF THYROID
CANCEFUCHARLESAND-
I Surgical
Pathology
FNAB
Papillary Cancer
Class 1 Class 2 Class 3 Class 4 Nondiagnostic Total
6 5 30 45 3 89
TABLE
on Fine Needle
Papillary Follicular
(4%) (12%) (20%) (83%) (14%) (21%)
4 1 14 5 2 26
Cell Type
Follicular Papillary Hikthle Not specified Total
(3%) (2%) (9%) (9%) (9%) (6%)
1 2 6 0 0 9
on Fine Needle
Aspiration
Class
I
2
Class
95 (7%) 0 11 (0%) 44 (9%) 150
(FNAB) Hiirthle Cancer
(1%) (5%) (4%)
0 2 6 1 0 9
(2%)
of patients
JOURNAL
OF SURGERY”
VOLUME
174
Biopsy Class
33 (15%) 0 9 (44%) 1 (100%) 43
diagnostic modality of choice in the management of thyroid nodules. Fine needle aspiration biopsy is quite accurate in distinguishing cancer from benign disease. Reported false-negative rates range from 1% to 11% and false-positive rates from 1% to 8%.’ In this study, we report a false-negative rate of 7% and a false-positive rate of 6% which is well within these ranges. The number of nondiagnostic aspirates in this series was 5%, which compares favorably with the reported rates of 2% to 21%.’ This low incidence may be due to the fact that repeat aspirates were frequently performed if the initial FNAB was nondiagnostic. Giuffrida and Gharib13 reported that 64% to 73% of aspirates were benign, 3% to 4% malignant, and 11% to 17% suspicious or indeterminate. In our series, 36% of FNABs are benign, 13% malignant, and 46% indeterminate or suspicious. The discrepancy in these figures when compared with other reported series arises from the fact that most patients in this report were referred after evaluation by their primary physician or endocrinologist suggested the need for surgery. Many had positive FNABs at the time of referral. Thirty-four percent of patients with an indeterminate biopsy (class 2 or 3) were found to have cancer at surgery compared with the reported average of 30%.4,5 Biopsies with cells showing papillary differentiation but not meeting strict criteria for malignancy were associated with a diagnosis of cancer in 56% of patients. For that reason, we recommend that all patients whose FNABs show papillary features undergo surgery. In the remaining patients with indeterminate nonpapillary cytology, the incidence of malignancy was 29%. We attempted to study other factors that might be associated with an increased risk of malignancy in this subgroup of patients. Neither age, gender, cell type, nor tumor size correlated with an increased incidence of malignancy. Because the majority of patients in the indeterminate group who undergo surgery are found to have benign disTHE AMERICAN
Biopsy
Follicular Cancer
Type
Numbers in parentheses indicate percentage Twenty-two FNABs were nondiagnostic.
546
Aspiration
Cancer Variant
Benign 139 33 97 3 17 289
(5%) (4%) (2%) (2%)
(93%) (77%) (63%) (6%) (77%) (68%)
II Cell
FNAB
Based
57 27 52 17 153
(FNAB) 3
(35%) (56%) (23%) (41%)
Class 0 54 (94%) 0 0 54
4
Total 185 81 72 62 400
with mct?r,
ease, other authors have tried to identify factors that could further subdivide this group and predict the likelihood of thyroid cancer. Tyler et all4 discussed the management of patients with indeterminate FNABs. In their study, 104 patients with indeterminate cytology were categorized into four groups based on clinical suspicion of cancer; 81 patients underwent thyroidectomy. The majority of invasive cancers were found in patients whose lesions had been classified as suspicious for papillary cancer. In addition, patients over 50 years of age who had a cytologic diagnosis of follicular or Hiirthle cell neoplasm were also identified as a subgroup with increased risk for cancer. Hamming et al’ evaluated the accuracy of preoperative FNAB by dividing patients into high, moderate, and low categories of suspicion on clinical grounds without previous knowledge of either the histologic or cytologic result. Clinical factors including rapid growth, nodule size, age, and radiation history were used to assign patients to each group. The rate of malignancy in the high suspicion group was 71% whereas the rate in the moderate and low suspicion groups were 14% and ll%, respectively. The presence of follicular cells in an FNAB may indicate a neoplasm in 10% to 60% of patients4 In this study, 17% of patients with follicular cells in the FNAB were shown to have cancer. Similarly, only 22% of patients with Htirthle cells in this series had cancer. This suggests that the finding of follicular cells or Hiirthle cells should not necessarily be considered an indication for surgery. Furthermore, most patients with follicular cells on FNAB who have cancer have papillary rather than follicular cancer. Based on cytologic criteria alone, we attempted to separate patients whose aspirates were suspicious (class 3) from those that were only slightly suspicious (class 2). The difference in the incidence of cancer in these two groups is not significant, suggesting that based on current cytologic criteria it is not possible to stratify the degree of suspicion once those patients with papillary cytology are excluded. NOVEMBER
1997
1 CYTOLOGIC
REFERENCES 1. Caruso D, Mazzaferri EL. Fine needle aspiration biopsy in the management of thyroid nodules. Endocrinologist. 1991;1:194202. 2. Hamming JF, Goslings BM, vanSteenis 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. Arch Intern Med. 1990;150:113-116. 3. Gharib H. Fine needle aspiration biopsy of thyroid nodules: advantages, limitations and effect. Mayo Clin Proc. 1994;69:44-49. 4. DeJong SA, Demeter JG, Castelli M, et al. Follicular cell predominance in the cytologic examination of dominant thyroid nodules indicates a sixty percent incidence of neoplasia. Surgery. 1990;108:794-799. 5. Gharib H, Goellner JR, Johnson DA. Fine needle aspiration cy tology of the thyroid. Clin Lab Med. 1993;13:699-709. 6. Mazzaferri EL. Management of a solitary thyroid nodule. NEJM. 1993;328:189-195. 7. Gharib H, Goellner JR. Fine needle aspiration biopsy of the thy raid: an appraisal. Ann Intern Med. 1993;118:282-289.
THE
AMERICAN
DETERMINANTS
OF THYROID
CAiCERKHARLES
AND
HELLER~:
8. Martin HE, Ellis EB. Biopsy by needle puncture and aspiration. Ann Surg. 1930;92:169-181. 9. Miller TM, Hamburger ]I, Kim SR. The impact of needle biopsy on the preoperative diagnosis of thyroid nodules. Henry Fcxd Hasp MedJ. 1980;28:145-148. 10. Rosen IB, Wallace C, Strawridge HG, et al. Reevaluation of needle aspiration cytology in detection of thyroid cancer. Surgery. 1981;90:747-756. 11. Colacchio TA, LoGerfo P, Feind LR. Fine needle cytologic diagnosis of thyroid nodules. Review and report of 300 cases. Am J Surg. 1980;149:568-571. 12. Hamberger B, Gharib H, Melton LJ III, et al. Fine needle aspiration biopsy of thyroid nodules. Impact on thyroid practice and cost of care. Am J Med. 1982;73:381-384. 13. Giuffrida D, Gharib H. Controversies in the management of cold, hot, and occult thyroid nodules. Am J Med. 1995;99:642-650. 14. Tyler DS, Winchester DJ, Caraway NP, et al. Indeterminate fine-needle aspiration biopsy of the thyroid: identification of subgroups at high risk for invasive carcinoma. Surgery. 1994;16:10541060.
JOURNAL
OF
SURGERY@
VOLUME
174
NOVEMBER
1997
547