- Email: [email protected]
Thyroid cancer
Thyroid Cancer Twenty-Three Years’ Experience at Baptist and St. Thomas Hospitals
John A. Shields, MD, Nashville, Tennessee John L. Farringer, Jr, MD, Nashville, Tennessee
The first true extirpation of the thyroid recorded was performed in the year 330 by Albucasis, an Arab court physician living in Baghdad [I]. In 1868, William Warren Green [2] of Portland, Maine performed the first successful thyroidectomy, although medical historians gave the credit for this to Theodor Kocher ten years later [3]. Kocher of Berne, Switzerland performed this operation more than 2,000 times with only a 4.5 per cent mortality. Another student of Billroth, Von Eiselsberg (1860), was perhaps the first to study thyroid cancer and its metastases [4]. Since Kocher’s time, hundreds of thousands of goiters, benign and malignant, have been removed. These numbers make it all the more remarkable that only in the past decade have the several types of cancer occurring in the thyroid gland and their individual biologic behavior attracted real attention. The lack of a uniformly accepted classification has led to confusion in evaluating malignant tumors of the thyroid, and in our series there was often a hesitancy to make a definitive cellular diagnosis. Material
During the period from 1952 to 1975,2,711 thyroidectomies were performed at Baptist and St. Thomas Hospitals. The charts of 156 of these patients with a diagnosis of carcinoma of the thyroid have been reviewed. Post-hospital
From the Department of Surgery, Baptist Hospital, Nashville, Tennessee. Reprint requests shoukl be addressed to John A. Shiikfs, M), 2000 Church Street, Nashville, Tennessee 37236.
Volume 133, February 1977
follow-up obtained in ninety-six patients is sufficient to evaluate their ultimate course and forms the basis of this report. Carcinomas are classified according to the classification of de Groot [5].
Results
Our group of 156. patients included 59 patients (37.8 per cent) with papillary carcinoma, 37 (23.7 per cent) with follicular carcinoma (including 9 with Hurthle cell carcinoma), 35 with mixed carcinoma, l(0.64 per cent) with medullary carcinoma, 20 (12.8 per cent) with anaplastic cancers, 1 (0.64 per cent) with lymphosarcoma, 1 (0.64 per cent) with fibrosarcoma, and 2 (1.28 per cent) with metastatic tumor. The overall mortality rate with a ten to twenty year follow-up was 15 per cent. (Figure 1.) Patients with papillary cancer had the most favorable relative prognosis and showed a 9 per cent mortality rate. This increased slightly in patients with mixed papillary and follicular carcinoma to 11.4 per cent. A more guarded relative prognosis was noted in patients with follicular carcinoma (including Hurthle cell carcinoma), with a mortality of 23.9 per cent. Anaplastic carcinomas occurred almost entirely in older patients, with a peak incidence in the sixth decade, and mortality here approached 100 per cent. (Figure 2.) Thyroid carcinoma was diagnosed more commonly in women (111 females, 45 males), the ratio being nearly equal in the fifth decade in our series. (Figure
211
Shields and Farringer
0 Medullary
(Icase) Papillary (59 cases) ._-- .____o._b Mixed (35 cases1
l-•2_.---._. Follicular
2 k?
5
(37 cases)
i 50
i
-
i
5 (D aQ
i
\ i 25-
i, ‘\ ‘\
0
.‘\‘A Anaplastic
(20
cases)
I
I
I
I
I
5
IO
I5
20
25
YEARS
FOLLOW-UP
Figure 1. Survival rates by histologic type in patients with thyroid carcinoma.
x
TOTAL CASES
30 t
l-l 20
30
40
SO
60
70
80
90
Figure 2. Age and sex distribution among patients presenting- with thyroid carcinoma.
3.) Age and sex significantly affected survival in our series; the female ten and twenty year survival figures were better than those achieved at five years in the male. (Figure 4.) Survival figures by age at diagnosis show a significant influence also. Patients with thyroid cancer presenting after sixty years of age fare far worse than those with cancer occurring earlier. This applies equally to both sexes. The actual histologic pattern appears to have little bearing on the course of the disease, which depends to a remarkable degree on the histologic grade and invasiveness of the tumor. In our series patients with well encapsulated papillary and follicular cancers were cured by excision of the involved lobe. Nineteen patients with cervical node metastases were treated by total thyroidectomy or lobectomy combined with lymph node dissection; carcinoma has reappeared in the neck or mediastinum in only six of these, with three deaths. The recurrence and mortality after lesser operative procedures carried out during the same period of time were considerably higher. Twenty-three patients with clinically positive nodes underwent thyroidectomy without node dissection and twenty-one of these are now dead. One hundred fourteen patients did not have clinical evidence of cervical node metastases, but elective neck dissection was performed in six of these. One had occult metastases in the dissected lymph nodes and subsequently had other metastases. Four toxic goiters were found to contain carcinoma. Three of these were removed surgically while the fourth was discovered on postmortem examination.
<30
(13 Cores)
A=-----•--=-•--=*
I
-A-k._
1
, YEARS
SEX
No.
of CASES
I5
IO
5
‘A-. ‘.
.A..
)60 -.-A_.
(19 Cases) -+.-.-.
I
1
I
,
5
IO
15
20
20
YEARS
FOLLOW-UP
FOLLOW-UP
5 YEARS
% SURVIVAL IO YEARS
20 YEARS
Femole
76
63%
56%
44%
Male
20
35%
25%
20%
Figure 3. Survival by sex in ninety-six follow-up patients with thyroid carcinoma.
212
‘.
AGE AT DIAGNOSIS (30 30-60 )60
No. of CASES
5 YEARS
!4 I9
100% 81% 32%
% SURVIVAL IO YEARS 100% 72% 21%
20 YEARS 100% 66% 10%
Figure 4. Survival by age in ninety-six follow-up patients with thyroid carcinoma.
The American
Journal
of Surgery
Thyroid Cancer
Slightly more than 50 per cent of our series presented with multinodular goiters. Patients of both sexes have a greater mortality from thyroid cancer with advancing age. However, it has been shown that females are selectively protected until they reach the menopausal period. Their good prognosis overall is related to this. Comments
The identification of malignancy in goiters has always been a problem to surgeons. Criteria helping to distinguish benign from malignant nontoxic multinodular goiters have been defined in the belief that removal of all goiters would result in more lives lost than saved [6, 71. Although thyroid disease is more common in women in general, when a man has a nontoxic nodular goiter, it is three times more likely that he will have cancer of the thyroid. Consequently, suspicion of cancer of the thyroid should have top priority in these patients. A decision not to operate on a nontoxic nodular goiter can only be made in the knowledge that carcinoma has been absolutely excluded. These patients being seen by the surgeon should therefore be regarded as having been selected for surgery. A realistic approach is the employment of bilateral subtotal thyroidectomy with conversion to total thyroidectomy with neck dissection if cancer is found or suspected in cervical nodes. In our series, total thyroidectomy was seldom used in the treatment of differentiated carcinomas and probably accounts for poor results obtained in some cases with multicentric carcinomas. Multicentricity has been found to be a common feature especially of papillary tumors in children. Estimates range from 20 to 80 per cent of proved multicentricity, depending on the perseverance of the pathologist in studying the tissue sections. Whether these are truly multicentric sites of neoplasia or intrathyroidal metastases is not clear [8]. Liechty, Safaie-Shirazi, and Soper [9] reported that 30 per cent of their patients had multicentric foci of cancer whereas Clark, White, and Russell [IO] and Tawes and de Lorimier [I I] reported multicentricity in more than 80 per cent of their histologic material. In the majority of patients younger than eighteen years the first symptom was a slow growing cervical nodule, which early created little suspicion and often was regarded as chronic adenitis or a branchial cleft cyst. This is reflected in the elapsed time between discovery and time of referral, which ranged up to six years [9].
Volume 133, February 1977
Crile [12] suggested that a high index of suspicion should be directed at all solitary nodules and was able to report a 90 per cent success rate in diagnosing carcinoma preoperatively in solitary nodules. In contrast, Hoffman et al [13] state that there is no way to differentiate benign from malignant solitary nodules and, therefore, all must be resected, a view previously expressed by Liechty, Graham, and Freemeyer [14]. Occasionally, either papillary or follicular tumors may appear as small lesions with a surrounding dense fibrotic reaction. These sclerosing carcinomas, however, are known to metastasize and appear to be as malignant as those without a capsule [15]. The carcinoma incidence of 0.25 per cent in excised toxic goiters in our series is similar to the 0.34 per cent of Hurxthal[16] and the 0.4 per cent of Campbell and Sage [17]. The greatest controversy in the surgical treatment of carcinoma of the thyroid, especially papillary is whether routinely to perform some type of neck dissection. Noting the high incidence of positive nodes in specimens taken at elective neck dissection for cancer of the thyroid, ranging from 21 to 82 per cent, many surgeons prefer elective standard or modified neck dissection even when nodes are not palpable. Block, Miller, and Horn [18] reported on thirty-five patients without preoperative evidence of cervical lymphadenopathy for whom cervical node dissection was done. Nineteen (54 per cent) were found to have microscopic evidence of metastases to the nodes. In forty-two of their patients treated by total thyroidectomy and lymph node dissection, lymphadenopathy was the initial indication of carcinoma in fifteen. Cervical nodes were enlarged preoperatively in twenty-eight patients in the selected group. A few patients in our series underwent reoperation for enlarged nodes located lateral to the posterior triangle and found to be free of disease. Tollefsen, DeCosse, and Hutter [19] found that of sixty-nine patients who had elective neck dissections, only two died of thyroid cancer and both had pulmonary metastases when first seen. Block, Miller, and Horn [3] found metastases to lymph nodes in the anterosuperior mediastinum in 9 per cent of their 284 patients and believed that they could be removed by an appropriate dissection through the cervical incision. McClintock et al [ZO] have advised a sternumsplitting approach to this dissection. It appears, therefore, that the presence of palpable enlarged cervical lymph nodes preoperatively nearly always indicates the presence of metastases from an existing thyroid carcinoma, but absence of palpable
213
Shields and Farringer
nodes does not preclude the existence of metastases. Of greater importance is the assessment at the time of operation of lymph nodes along the recurrent laryngeal nerves, in the anterosuperior mediastinum, and along the internal jugular vein. Failure to find node metastases here at operation indicates that the later appearance of cervical node spread is unlikely. Under such circumstances prophylactic lymph node dissection appears unnecessary, according to Block, Miller, and Horn [18]. It is believed that the patients with possible or known metastatic thyroid carcinoma, particularly of the papillary or follicular type, should receive a satisfactory radioiodine study to establish the functional status of the tumor. If the lesions are functional, then the possibility of treatment with radioactive iodine could well be entertained, with the possibility of avoiding numerous surgical procedures [21]. In our series, of thirty-seven follicular carcinomas, nine were classified as being Hurthle cell or clear cell carcinomas. In several of these complete surgical excision was not possible and employment of radical radiotherapy was essential. Despite the resultant thyroid destruction, some of these patients were not given thyroxine. These hormone-dependent tumors, therefore, were exposed to high levels of thyroidstimulating hormone (TSH). Three patients with papillary carcinoma in our series have survived for a long period of time on large doses of thyroid hormone. The evidence for use of thyroid suppression in the long-term management of patients with well differentiated thyroid cancer stems from experimental as well as clinical observations. In the experimental animal, TSH has been shown to serve as a promoting factor in the genesis of thyroid neoplasms. In the absence of such a stimulus, few neoplasms develop. There are a number of clinical observations after suppressive therapy for metastatic thyroid cancer with temporary, and at times apparently permanent, regression of the neoplasm. Furthermore, there is also evidence that in the presence of continued TSH stimulus (hypothyroidism), there may be progression of well differentiated to poorly differentiated thyroid carcinoma. This occurred in two of our patients. There are also unconfirmed data that TSH levels in patients with thyroid cancer may be elevated and that TSH in such patients may be more difficult to suppress. Studies of adenyl cyclase response of thyroid tumors in vitro show a marked receptivity and stimulation by TSH [16]. On the other hand, Sisson and Beierwaltes [7] showed that in the Fisher rat with papillary and follicular carcinoma of the thyroid, thyroidectemized
214
animals experienced a more rapid growth of the carcinoma when placed on thyroid hormone replacement. Finally, there are no controlled studies on the efficacy of thyroid hormone and because of the long natural history of the disease there may never be. Conclusions
Of nineteen patients with thyroid carcinoma and cervical node metastases treated from 1952 through 1975 with total thyroidectomy or lobectomy and lymph node dissection, carcinoma has reappeared in the neck or mediastinum in six patients with three deaths. This is considerably less than the recurrence and mortality after lesser operative procedures carried out during the same period of time in patients with evidence of lymph node metastases. In our series enlarged lymph nodes located lateral to the posterior triangle of the neck do not contain carcinoma. The prognosis of any cancer is determined by its histologic type and clinical stage primarily. Good results obtained in treating well differentiated thyroid cancer are not reflected in the treatment of anaplastic growths. During the past two decades the surgical management of thyroid cancer in our two institutions has been more radical than previously and a policy has evolved in which the magnitude of operation is determined by the gross findings of the extent of the carcinoma at the time of the procedure and, to a lesser degree, by the histologic variety of the lesion. Anaplastic carcinomas were almost uniformly fatal in our series. Even though mixed thyroid carcinoma is a distinct histologic entity, the clinical course of these patients is similar to that of patients with papillary carcinoma and they should be so treated. Hurthle cell cancers, although classified as a separate entity, were actually a morphologic variant of follicular carcinoma and survival and cure rates are comparable. There are no prospective randomized clinical studies of the effectiveness of various modes of surgical treatment. Summary
One hundred fifty-six patients with thyroid cancer were diagnosed and treated at Baptist and St. 0Thomas Hospitals from 1952 through 1975. Papillary carcinoma, representing 65 to 70 per cent of the total, occurs in all age groups and is the most readily curable of all carcinomas. Slightly more than 50 per cent
The American Journal of SurgOry
Thyroid Cancer
of our series presented with multinodular goiters. When a male patient has nontoxic nodular goiter, it is three times more likely that he will have cancer of the thyroid than a female patient with a similar goiter. Definitive thyroid surgery was performed by over fifty different surgeons in 143 patients and simultaneous neck dissection in twenty-five. The histologic types ranged from papillary (9 per cent mortality) to anaplastic carcinoma (nearly 100 per cent mortality). Age and sex were shown in our series to affect survival. The female survival figures were better than the male, and older patients fared far worse then younger ones. Survival rates are much improved in patients with cervical node metastases when radical neck dissection is done. Patients reoperated for enlarged nodes located lateral to posterior triangle were found not to have metastatic cancer. Hyperthyroidism was confirmed in 0.25 per cent. References 1. Beahrs OH, Tachovsky FJ: Surgery of the thyroid gland, chapt 1, p 1. Head and Neck Surgery. New York, Harper 8 Row, 1974. 2. Attie JN, Khafif RA, Steckler RM: Elective neck dissection in papillary carcinoma of the thyroid. Am J Surg 122: 469, 1971. 3. Block MA, Miller JM. Horn RC Jr: Significance of mediastinal lymph node metastases in carcinoma of the thyroid. Am J Surg 123: 704, 1972. 4. Beahrs OH: Surgery of the thyroid gland. Head Neck Surg 11:
Volume 133, February 1977
2, 1974. 5. DeGroot TJ: Thyroid carcinoma. Med C/in North Am 59: 5, 1975. 6. Hoffman GI, Thompson NW, Heffron C: The solitary thyroid nodule. Arch Surg 195: 379, 1972. 7. Sisson JC, Belerwaltes WH: Effect of thyroidectomy with and without thyroxine replacement on transplantable thyroid tumor in rats. Endocrinology 74: 925, 1964. 6. Russell W, lbanex M. Clark R: Thyroid carcinoma. Cancer 16: 425, 1963. 9. Liechty RD, Safale-Shirazi S, Soper RT: Carcinoma of the thyroid in children. Surg Gynecol Obsfet 134: 595, 1972. 10. Clark RT, White EC, Russell WO: Total thyroidectomy of cancer of the thyroid; significance of intraglandular dissemination. Ann Surg 149: 858, 1959. 11. Tawes RT, DeLorimier AA: Thyroid carcinoma during youth. J Pediatr Surg 3: 210, 1968. 12. Crlle G: Adenoma and carcinoma of the thyroid gland. N fngi J Med 249: 585, 1953. 13. Garrison FH: History of Medicine, 4th ed. Philadelphia, WB Saunders, 1960, p 807. 14. Liechty RD, Graham H, Freemeyer P: Benign solitary thyroid nodules. Surg Gynecol Obstet 121: 571, 1965. 15. Crile G: Endocrine dependency of papillary carcinoma of the thyroid. JAMA 195: 721, 1966. 16. Decosse JJ, Beierwaltes WH, Brooks JR, Thomas CG Jr, Woo&r LB: Carcinoma of the thyroid. Arch Surg 110: 783, 1975. 17. Campbell DJ, Sage RH: Thyroid cancer: twenty years experience in a general hospital. Br J Surg 62: 207, 1975. 18. Block MA, Miller JM, Horn RC Jr: Thyroid cancer with cervical lymph node metastases. Am J Surg 122: 458, 197 1. 19. Tollefsen HR. DeCosse JJ, Hutter RV: Papillary carcinoma of the thyroid. A clinical and pathological study of 70 fatal cases. Cancer 17: 1035, 1964. 20. McClintock JC, Stranahan T, Alley RD, Baker WA: A thoracocervical approach to malignant disease of the thyroid gland. Ann Surg 139: 158, 1954. 21. Wellman HN: Metastatic thyroid carcinoma. JAMA 227: 439, 1974.
215
John A. Shields, MD, Nashville, Tennessee John L. Farringer, Jr, MD, Nashville, Tennessee
The first true extirpation of the thyroid recorded was performed in the year 330 by Albucasis, an Arab court physician living in Baghdad [I]. In 1868, William Warren Green [2] of Portland, Maine performed the first successful thyroidectomy, although medical historians gave the credit for this to Theodor Kocher ten years later [3]. Kocher of Berne, Switzerland performed this operation more than 2,000 times with only a 4.5 per cent mortality. Another student of Billroth, Von Eiselsberg (1860), was perhaps the first to study thyroid cancer and its metastases [4]. Since Kocher’s time, hundreds of thousands of goiters, benign and malignant, have been removed. These numbers make it all the more remarkable that only in the past decade have the several types of cancer occurring in the thyroid gland and their individual biologic behavior attracted real attention. The lack of a uniformly accepted classification has led to confusion in evaluating malignant tumors of the thyroid, and in our series there was often a hesitancy to make a definitive cellular diagnosis. Material
During the period from 1952 to 1975,2,711 thyroidectomies were performed at Baptist and St. Thomas Hospitals. The charts of 156 of these patients with a diagnosis of carcinoma of the thyroid have been reviewed. Post-hospital
From the Department of Surgery, Baptist Hospital, Nashville, Tennessee. Reprint requests shoukl be addressed to John A. Shiikfs, M), 2000 Church Street, Nashville, Tennessee 37236.
Volume 133, February 1977
follow-up obtained in ninety-six patients is sufficient to evaluate their ultimate course and forms the basis of this report. Carcinomas are classified according to the classification of de Groot [5].
Results
Our group of 156. patients included 59 patients (37.8 per cent) with papillary carcinoma, 37 (23.7 per cent) with follicular carcinoma (including 9 with Hurthle cell carcinoma), 35 with mixed carcinoma, l(0.64 per cent) with medullary carcinoma, 20 (12.8 per cent) with anaplastic cancers, 1 (0.64 per cent) with lymphosarcoma, 1 (0.64 per cent) with fibrosarcoma, and 2 (1.28 per cent) with metastatic tumor. The overall mortality rate with a ten to twenty year follow-up was 15 per cent. (Figure 1.) Patients with papillary cancer had the most favorable relative prognosis and showed a 9 per cent mortality rate. This increased slightly in patients with mixed papillary and follicular carcinoma to 11.4 per cent. A more guarded relative prognosis was noted in patients with follicular carcinoma (including Hurthle cell carcinoma), with a mortality of 23.9 per cent. Anaplastic carcinomas occurred almost entirely in older patients, with a peak incidence in the sixth decade, and mortality here approached 100 per cent. (Figure 2.) Thyroid carcinoma was diagnosed more commonly in women (111 females, 45 males), the ratio being nearly equal in the fifth decade in our series. (Figure
211
Shields and Farringer
0 Medullary
(Icase) Papillary (59 cases) ._-- .____o._b Mixed (35 cases1
l-•2_.---._. Follicular
2 k?
5
(37 cases)
i 50
i
-
i
5 (D aQ
i
\ i 25-
i, ‘\ ‘\
0
.‘\‘A Anaplastic
(20
cases)
I
I
I
I
I
5
IO
I5
20
25
YEARS
FOLLOW-UP
Figure 1. Survival rates by histologic type in patients with thyroid carcinoma.
x
TOTAL CASES
30 t
l-l 20
30
40
SO
60
70
80
90
Figure 2. Age and sex distribution among patients presenting- with thyroid carcinoma.
3.) Age and sex significantly affected survival in our series; the female ten and twenty year survival figures were better than those achieved at five years in the male. (Figure 4.) Survival figures by age at diagnosis show a significant influence also. Patients with thyroid cancer presenting after sixty years of age fare far worse than those with cancer occurring earlier. This applies equally to both sexes. The actual histologic pattern appears to have little bearing on the course of the disease, which depends to a remarkable degree on the histologic grade and invasiveness of the tumor. In our series patients with well encapsulated papillary and follicular cancers were cured by excision of the involved lobe. Nineteen patients with cervical node metastases were treated by total thyroidectomy or lobectomy combined with lymph node dissection; carcinoma has reappeared in the neck or mediastinum in only six of these, with three deaths. The recurrence and mortality after lesser operative procedures carried out during the same period of time were considerably higher. Twenty-three patients with clinically positive nodes underwent thyroidectomy without node dissection and twenty-one of these are now dead. One hundred fourteen patients did not have clinical evidence of cervical node metastases, but elective neck dissection was performed in six of these. One had occult metastases in the dissected lymph nodes and subsequently had other metastases. Four toxic goiters were found to contain carcinoma. Three of these were removed surgically while the fourth was discovered on postmortem examination.
<30
(13 Cores)
A=-----•--=-•--=*
I
-A-k._
1
, YEARS
SEX
No.
of CASES
I5
IO
5
‘A-. ‘.
.A..
)60 -.-A_.
(19 Cases) -+.-.-.
I
1
I
,
5
IO
15
20
20
YEARS
FOLLOW-UP
FOLLOW-UP
5 YEARS
% SURVIVAL IO YEARS
20 YEARS
Femole
76
63%
56%
44%
Male
20
35%
25%
20%
Figure 3. Survival by sex in ninety-six follow-up patients with thyroid carcinoma.
212
‘.
AGE AT DIAGNOSIS (30 30-60 )60
No. of CASES
5 YEARS
!4 I9
100% 81% 32%
% SURVIVAL IO YEARS 100% 72% 21%
20 YEARS 100% 66% 10%
Figure 4. Survival by age in ninety-six follow-up patients with thyroid carcinoma.
The American
Journal
of Surgery
Thyroid Cancer
Slightly more than 50 per cent of our series presented with multinodular goiters. Patients of both sexes have a greater mortality from thyroid cancer with advancing age. However, it has been shown that females are selectively protected until they reach the menopausal period. Their good prognosis overall is related to this. Comments
The identification of malignancy in goiters has always been a problem to surgeons. Criteria helping to distinguish benign from malignant nontoxic multinodular goiters have been defined in the belief that removal of all goiters would result in more lives lost than saved [6, 71. Although thyroid disease is more common in women in general, when a man has a nontoxic nodular goiter, it is three times more likely that he will have cancer of the thyroid. Consequently, suspicion of cancer of the thyroid should have top priority in these patients. A decision not to operate on a nontoxic nodular goiter can only be made in the knowledge that carcinoma has been absolutely excluded. These patients being seen by the surgeon should therefore be regarded as having been selected for surgery. A realistic approach is the employment of bilateral subtotal thyroidectomy with conversion to total thyroidectomy with neck dissection if cancer is found or suspected in cervical nodes. In our series, total thyroidectomy was seldom used in the treatment of differentiated carcinomas and probably accounts for poor results obtained in some cases with multicentric carcinomas. Multicentricity has been found to be a common feature especially of papillary tumors in children. Estimates range from 20 to 80 per cent of proved multicentricity, depending on the perseverance of the pathologist in studying the tissue sections. Whether these are truly multicentric sites of neoplasia or intrathyroidal metastases is not clear [8]. Liechty, Safaie-Shirazi, and Soper [9] reported that 30 per cent of their patients had multicentric foci of cancer whereas Clark, White, and Russell [IO] and Tawes and de Lorimier [I I] reported multicentricity in more than 80 per cent of their histologic material. In the majority of patients younger than eighteen years the first symptom was a slow growing cervical nodule, which early created little suspicion and often was regarded as chronic adenitis or a branchial cleft cyst. This is reflected in the elapsed time between discovery and time of referral, which ranged up to six years [9].
Volume 133, February 1977
Crile [12] suggested that a high index of suspicion should be directed at all solitary nodules and was able to report a 90 per cent success rate in diagnosing carcinoma preoperatively in solitary nodules. In contrast, Hoffman et al [13] state that there is no way to differentiate benign from malignant solitary nodules and, therefore, all must be resected, a view previously expressed by Liechty, Graham, and Freemeyer [14]. Occasionally, either papillary or follicular tumors may appear as small lesions with a surrounding dense fibrotic reaction. These sclerosing carcinomas, however, are known to metastasize and appear to be as malignant as those without a capsule [15]. The carcinoma incidence of 0.25 per cent in excised toxic goiters in our series is similar to the 0.34 per cent of Hurxthal[16] and the 0.4 per cent of Campbell and Sage [17]. The greatest controversy in the surgical treatment of carcinoma of the thyroid, especially papillary is whether routinely to perform some type of neck dissection. Noting the high incidence of positive nodes in specimens taken at elective neck dissection for cancer of the thyroid, ranging from 21 to 82 per cent, many surgeons prefer elective standard or modified neck dissection even when nodes are not palpable. Block, Miller, and Horn [18] reported on thirty-five patients without preoperative evidence of cervical lymphadenopathy for whom cervical node dissection was done. Nineteen (54 per cent) were found to have microscopic evidence of metastases to the nodes. In forty-two of their patients treated by total thyroidectomy and lymph node dissection, lymphadenopathy was the initial indication of carcinoma in fifteen. Cervical nodes were enlarged preoperatively in twenty-eight patients in the selected group. A few patients in our series underwent reoperation for enlarged nodes located lateral to the posterior triangle and found to be free of disease. Tollefsen, DeCosse, and Hutter [19] found that of sixty-nine patients who had elective neck dissections, only two died of thyroid cancer and both had pulmonary metastases when first seen. Block, Miller, and Horn [3] found metastases to lymph nodes in the anterosuperior mediastinum in 9 per cent of their 284 patients and believed that they could be removed by an appropriate dissection through the cervical incision. McClintock et al [ZO] have advised a sternumsplitting approach to this dissection. It appears, therefore, that the presence of palpable enlarged cervical lymph nodes preoperatively nearly always indicates the presence of metastases from an existing thyroid carcinoma, but absence of palpable
213
Shields and Farringer
nodes does not preclude the existence of metastases. Of greater importance is the assessment at the time of operation of lymph nodes along the recurrent laryngeal nerves, in the anterosuperior mediastinum, and along the internal jugular vein. Failure to find node metastases here at operation indicates that the later appearance of cervical node spread is unlikely. Under such circumstances prophylactic lymph node dissection appears unnecessary, according to Block, Miller, and Horn [18]. It is believed that the patients with possible or known metastatic thyroid carcinoma, particularly of the papillary or follicular type, should receive a satisfactory radioiodine study to establish the functional status of the tumor. If the lesions are functional, then the possibility of treatment with radioactive iodine could well be entertained, with the possibility of avoiding numerous surgical procedures [21]. In our series, of thirty-seven follicular carcinomas, nine were classified as being Hurthle cell or clear cell carcinomas. In several of these complete surgical excision was not possible and employment of radical radiotherapy was essential. Despite the resultant thyroid destruction, some of these patients were not given thyroxine. These hormone-dependent tumors, therefore, were exposed to high levels of thyroidstimulating hormone (TSH). Three patients with papillary carcinoma in our series have survived for a long period of time on large doses of thyroid hormone. The evidence for use of thyroid suppression in the long-term management of patients with well differentiated thyroid cancer stems from experimental as well as clinical observations. In the experimental animal, TSH has been shown to serve as a promoting factor in the genesis of thyroid neoplasms. In the absence of such a stimulus, few neoplasms develop. There are a number of clinical observations after suppressive therapy for metastatic thyroid cancer with temporary, and at times apparently permanent, regression of the neoplasm. Furthermore, there is also evidence that in the presence of continued TSH stimulus (hypothyroidism), there may be progression of well differentiated to poorly differentiated thyroid carcinoma. This occurred in two of our patients. There are also unconfirmed data that TSH levels in patients with thyroid cancer may be elevated and that TSH in such patients may be more difficult to suppress. Studies of adenyl cyclase response of thyroid tumors in vitro show a marked receptivity and stimulation by TSH [16]. On the other hand, Sisson and Beierwaltes [7] showed that in the Fisher rat with papillary and follicular carcinoma of the thyroid, thyroidectemized
214
animals experienced a more rapid growth of the carcinoma when placed on thyroid hormone replacement. Finally, there are no controlled studies on the efficacy of thyroid hormone and because of the long natural history of the disease there may never be. Conclusions
Of nineteen patients with thyroid carcinoma and cervical node metastases treated from 1952 through 1975 with total thyroidectomy or lobectomy and lymph node dissection, carcinoma has reappeared in the neck or mediastinum in six patients with three deaths. This is considerably less than the recurrence and mortality after lesser operative procedures carried out during the same period of time in patients with evidence of lymph node metastases. In our series enlarged lymph nodes located lateral to the posterior triangle of the neck do not contain carcinoma. The prognosis of any cancer is determined by its histologic type and clinical stage primarily. Good results obtained in treating well differentiated thyroid cancer are not reflected in the treatment of anaplastic growths. During the past two decades the surgical management of thyroid cancer in our two institutions has been more radical than previously and a policy has evolved in which the magnitude of operation is determined by the gross findings of the extent of the carcinoma at the time of the procedure and, to a lesser degree, by the histologic variety of the lesion. Anaplastic carcinomas were almost uniformly fatal in our series. Even though mixed thyroid carcinoma is a distinct histologic entity, the clinical course of these patients is similar to that of patients with papillary carcinoma and they should be so treated. Hurthle cell cancers, although classified as a separate entity, were actually a morphologic variant of follicular carcinoma and survival and cure rates are comparable. There are no prospective randomized clinical studies of the effectiveness of various modes of surgical treatment. Summary
One hundred fifty-six patients with thyroid cancer were diagnosed and treated at Baptist and St. 0Thomas Hospitals from 1952 through 1975. Papillary carcinoma, representing 65 to 70 per cent of the total, occurs in all age groups and is the most readily curable of all carcinomas. Slightly more than 50 per cent
The American Journal of SurgOry
Thyroid Cancer
of our series presented with multinodular goiters. When a male patient has nontoxic nodular goiter, it is three times more likely that he will have cancer of the thyroid than a female patient with a similar goiter. Definitive thyroid surgery was performed by over fifty different surgeons in 143 patients and simultaneous neck dissection in twenty-five. The histologic types ranged from papillary (9 per cent mortality) to anaplastic carcinoma (nearly 100 per cent mortality). Age and sex were shown in our series to affect survival. The female survival figures were better than the male, and older patients fared far worse then younger ones. Survival rates are much improved in patients with cervical node metastases when radical neck dissection is done. Patients reoperated for enlarged nodes located lateral to posterior triangle were found not to have metastatic cancer. Hyperthyroidism was confirmed in 0.25 per cent. References 1. Beahrs OH, Tachovsky FJ: Surgery of the thyroid gland, chapt 1, p 1. Head and Neck Surgery. New York, Harper 8 Row, 1974. 2. Attie JN, Khafif RA, Steckler RM: Elective neck dissection in papillary carcinoma of the thyroid. Am J Surg 122: 469, 1971. 3. Block MA, Miller JM. Horn RC Jr: Significance of mediastinal lymph node metastases in carcinoma of the thyroid. Am J Surg 123: 704, 1972. 4. Beahrs OH: Surgery of the thyroid gland. Head Neck Surg 11:
Volume 133, February 1977
2, 1974. 5. DeGroot TJ: Thyroid carcinoma. Med C/in North Am 59: 5, 1975. 6. Hoffman GI, Thompson NW, Heffron C: The solitary thyroid nodule. Arch Surg 195: 379, 1972. 7. Sisson JC, Belerwaltes WH: Effect of thyroidectomy with and without thyroxine replacement on transplantable thyroid tumor in rats. Endocrinology 74: 925, 1964. 6. Russell W, lbanex M. Clark R: Thyroid carcinoma. Cancer 16: 425, 1963. 9. Liechty RD, Safale-Shirazi S, Soper RT: Carcinoma of the thyroid in children. Surg Gynecol Obsfet 134: 595, 1972. 10. Clark RT, White EC, Russell WO: Total thyroidectomy of cancer of the thyroid; significance of intraglandular dissemination. Ann Surg 149: 858, 1959. 11. Tawes RT, DeLorimier AA: Thyroid carcinoma during youth. J Pediatr Surg 3: 210, 1968. 12. Crlle G: Adenoma and carcinoma of the thyroid gland. N fngi J Med 249: 585, 1953. 13. Garrison FH: History of Medicine, 4th ed. Philadelphia, WB Saunders, 1960, p 807. 14. Liechty RD, Graham H, Freemeyer P: Benign solitary thyroid nodules. Surg Gynecol Obstet 121: 571, 1965. 15. Crile G: Endocrine dependency of papillary carcinoma of the thyroid. JAMA 195: 721, 1966. 16. Decosse JJ, Beierwaltes WH, Brooks JR, Thomas CG Jr, Woo&r LB: Carcinoma of the thyroid. Arch Surg 110: 783, 1975. 17. Campbell DJ, Sage RH: Thyroid cancer: twenty years experience in a general hospital. Br J Surg 62: 207, 1975. 18. Block MA, Miller JM, Horn RC Jr: Thyroid cancer with cervical lymph node metastases. Am J Surg 122: 458, 197 1. 19. Tollefsen HR. DeCosse JJ, Hutter RV: Papillary carcinoma of the thyroid. A clinical and pathological study of 70 fatal cases. Cancer 17: 1035, 1964. 20. McClintock JC, Stranahan T, Alley RD, Baker WA: A thoracocervical approach to malignant disease of the thyroid gland. Ann Surg 139: 158, 1954. 21. Wellman HN: Metastatic thyroid carcinoma. JAMA 227: 439, 1974.
215