Moley, Jeffrey F. M.D.*†; Dilley, William G. Ph.D.*; DeBenedetti, Mary K. R.N.*
Objective: The purpose of the study is to determine whether reoperation for medullary thyroid carcinoma (MTC), performed with low morbidity in carefully selected patients, consistently results in improvement as determined by lowering of stimulated calcitonin levels.
Background: Persistent or recurrent elevation of stimulated plasma calcitonin levels occurs in >50% of patients after primary operation for MTC. Success of reoperation with clearance of metastatic cervical nodal disease has been hampered by failure to identify patients with distant metastases and by inadequate removal of involved nodal groups.
Methods: Since 1992, the authors have evaluated 115 patients with recurrent or residual MTC. Fifty-three patients have not undergone operation because of extent of disease, previous extensive treatment, medical condition, or patient choice. Sixty-two patients underwent surgery. Ten patients had laparoscopic or open examination of the liver, the results of which showed liver metastases. Seven patients had palliative debulking of cervical tumor. In 45 patients without evidence of distant metastases, cervical operation was carried out with curative intent. Removal of central, upper mediastinal, and lateral nodes (levels II, III, IV, VI, and VII) was done.
Results: Seven of eight patients who had palliative resections are alive without symptoms. In patients who underwent curative resections, postoperative stimulated calcitonin levels were in the normal range in 17 patients (38%) and were not significantly lowered in 6 patients (13%). There were no deaths, and no transfusions were used.
Conclusions: These results are a significant improvement over the authors' previous series and reflect better preoperative identification of patients with disease confined to the neck and improved operative strategy based on knowledge of the pattern of nodal spread of MTC.
Medullary thyroid carcinoma (MTC) is a moderately aggressive thyroid malignancy derived from neuroendocrine cells called C cells. These cells produce calcitonin, which is an excellent marker for the presence of MTC. Measurement of plasma calcitonin after the administration of calcium and pentagastrin is the most sensitive clinical test for the presence of MTC.1 Medullary thyroid carcinoma may occur without any known inherited genetic defect (sporadic MTC) or as a component of the hereditary cancer syndromes multiple endocrine neoplasia (MEN) type-2A, MEN-2B, and familial, non-MEN MTC.2
More than 50% of patients with MTC who present with a mass in the neck have nodal metastases, which are associated with a poor prognosis.3-5 Unfortunately, most patients with MTC are treated initially with total or subtotal thyroidectomy, and rarely is an adequate node dissection done. More than 50% of these patients have persistent disease after primary surgery, as evidenced by elevation of stimulated calcitonin levels.4,6 Unlike papillary thyroid cancer, however, in which residual disease in the neck and even distant metastases can be treated effectively with radioactive iodine, MTC does not take up iodine and is resistant to radiation therapy. An aggressive surgical approach, with complete thyroidectomy and systematic node clearance, is required for optimal results.7 Several groups have recognized the potential impact of reoperation for persistent cervical disease.6,8-13 Tisell et al.9 in 1986 published a series of 11 patients with persistent elevation of calcitonin levels after primary surgery. They performed an operation that they called a "microdissection," which entailed meticulous removal of central and lateral cervical nodes associated with the thyroid. They reported normalization of calcitonin levels in four patients. In 1993, our group reported a series of 32 patients in whom this approach was applied.14 In 9 (28%) of 32 patients, stimulated calcitonin levels were in the normal range in the immediate postoperative period. Ten patients (31%), however, had no change in calcitonin levels. Since that time, we have sought to improve these results by better selection of patients likely to benefit from reoperation. We have achieved this by applying systematic metastatic work-up, including routine computed tomography (CT) or magnetic resonance imaging (MRI) of the neck, chest, and abdomen, routine liver laparoscopy, and selective venous catheterization in selected patients. In the current series, seven patients underwent palliative cervical debulking of MTC, and 45 patients underwent systematic cervical dissection with curative intent after metastatic work-up showed no evidence of distant metastatic disease. Of these patients, 17 had initial postoperative stimulated calcitonin levels in the normal range, and only 6 patients had no change in calcitonin levels.
Between August 1992 and November 1996, 115 patients with recurrent or persistent elevation of calcitonin levels after primary surgery for MTC were evaluated (Table 1). Of these 115 patients, 3 reside in the St. Louis area, and the remainder reside in other parts of the United States (106 patients), Europe (4 patients), and the Middle East (2 patients). The patients were either seen in the surgeon's office (70 patients), or they were evaluated through telephone interviews with the patient and the referring physician (45 patients). In all patients, previous operative notes and pathology reports were reviewed.
Fifty-three patients have not undergone surgery. Of these, ten were found to have distant metastases after radiologic evaluation. Three patients who had extensive prior surgery and radiation therapy to the neck were not thought to be appropriate candidates for reoperation. Two patients were not thought to be candidates for reoperation because of severe underlying medical problems. Twenty-four patients declined further evaluation and consideration for reoperation after discussion of treatment options with this surgeon and with the referring physician. Fourteen patients currently are being evaluated or are undecided about further treatment.
Sixty-two patients underwent surgery. In all these patients except one, pathology slides from patients' initial operation were obtained and reviewed. In ten patients, laparoscopic or open examination of the liver showed liver metastases. Further cervical dissection was not carried out in these patients. In seven patients, palliative debulking of cervical tumor was carried out (Table 2).
In 45 patients, 51 operations were carried out with curative intent. The mean age of these patients was 38.8 years (range, 10-60 years). Twenty-four patients had sporadic disease, 17 patients had MEN-2A, and 4 patients had MEN-2B. Selective venous catheterization was performed in 20 of these patients, as described previously.15 Computed tomography or MRI of the neck, chest, and abdomen was done in 32 patients. Nine patients had nuclear scanning. Routine laboratory work was obtained in all cases, including blood count, chemistries, liver function tests, and coagulation studies.
A calcium-pentagastrin stimulated calcitonin test was carried out at our institution immediately before the operation and 72 hours after the operation. In this test, a standard injection of intravenous calcium (2 mg/kg per minute) and pentagastrin (0.5 µg/kg per 5 seconds) is given. Calcitonin levels are drawn at 1, 3, and 5 minutes from a peripheral vein. Calcitonin levels were measured by radioimmunoassay (SB-1)1 until December 1994 and, after that time, by the commercially available Nichols assay. For 1- and 2-year follow-up testing, patients either returned to our institution, where a calcium-pentagastrin stimulated test was performed, or they were tested by a physician in their geographic area. In those instances, calcitonin levels were either obtained after calcium-pentagastrin stimulation, after pentagastrin-only stimulation, or after no stimulation.
Direct examination of the liver was done by laparoscopy in 44 patients and as an open procedure in 3 patients. Laparoscopy was carried out as described previously.15 If hepatic metastases were identified, cervical exploration was not carried out. If the liver was clear of apparent metastatic disease, cervical exploration was carried out (Fig. 1). In the 45 patients who underwent cervical operation with curative intent, direct examination of the liver was carried out immediately before the neck procedure, with the patient under the same anesthetic.
Our technique of cervical exploration is as follows. The patient is placed in the reflex position with padding under the arms and torso. The head is placed in mild hyperextension. The previous thyroidectomy incision is excised and extended to the posterior borders of the sternocleidomastoid muscles. Superior and inferior subplatysmal flaps are developed and retracted. Exploration of the central compartment was done in all 45 patients undergoing reoperation with curative intent. In this procedure, the anterior borders of the sternocleidomastoid muscles are mobilized, and the recurrent nerves are identified by a "back door" approach after medial reflection of the sternothyroid muscle. Once the nerve is identified, the strap muscles may safely be mobilized off of the central compartment. The strap muscles are divided on one side and occasionally on both sides to enhance exposure. All residual thyroid tissue and nodal tissue within the central compartment is removed (level VI and VII nodes). This is carried out from the level of the hyoid bone, along the larynx, and along the recurrent nerves to a level below the innominate artery on the right side, and to a comparable level behind the clavicle on the left side. All tissue on the anterior surface of the trachea, including any visible thymic tissue, is removed. The lateral boundaries of this part of the dissection are the carotid sheaths. In patients with functioning parathyroid tissue that has not been autotransplanted, a thorough search for parathyroid glands is carried out. This may take 1 or more hours, because no nodes are removed until parathyroid tissue is identified and confirmed by biopsy and frozen section. When normal parathyroid tissue is identified, it is either resected and preserved on ice for autotransplantation or left in situ. After histologic verification, autotransplantation is carried out by mincing the parathyroid into 1- × 3-mm fragments and placing two fragments into each of ten pockets in the sternocleidomastoid muscle or into the muscle of the nondominant forearm, as described previously.16
Bilateral (30 patients) or unilateral (15 patients) functional neck dissections also were carried out. In this dissection, mobilization of the carotid sheath and removal of all nodal tissue anterior to the carotid sheath up to the level of the hypoglossal nerve is done. Laterally, the posterior jugular chain of nodes (levels II, III, and IV) are excised routinely. In the absence of palpable adenopathy, we do not routinely do an extensive dissection of posterior triangle nodes (level V). The lateral boundary of the dissection is the cervical plexus as it comes around the posterior border of the sternocleidomastoid muscle. The recurrent nerves, phrenic nerves, sympathetic chain, accessory nerve, hypoglossal nerve, marginal mandibular nerve, thoracic duct, and major vascular structures are carefully preserved. A drain is placed in the central compartment, and the strap muscles are carefully reapproximated and tacked together in the midline. The platysma is closed and skin is closed with a running subcuticular suture. Postoperative care includes elevation of the head of the bed and prevention of coughing and vomiting, which increase venous pressure in the neck and can cause bleeding. A liquid diet is given on the night after surgery, and a regular diet is resumed the next day.
The variables were compared individually among the three groups, and differences were determined by Fisher's exact test. Data are presented as mean ± standard error of the mean.
The characteristics of the seven patients who had palliative cervical operations are summarized in Table 2. Five of these patients are alive and well with no evidence of airway compromise or of bulky cervical recurrence. One patient died with massive liver involvement after a prolonged period characterized by weight loss, diarrhea, and inanition.
Computed tomography or MRI of the neck was obtained in 32 of 45 patients. In four of these 32 patients, tumor was palpable in the neck. In five cases, CT or MRI showed cervical tumor that was not palpable. The results of 23 studies were negative, and in all of these patients, tumor was removed from the neck at operation. Selective venous catheterization correctly localized tumor to a general region of the neck in 17 of 20 cases.
Histologic examination of resected specimens confirmed removal of tumor in 43 patients. In two patients, there was no evidence of tumor on routine histologic examination, but both patients had a decrease in stimulated calcitonin levels to the normal range after surgery.
Figure 2 shows the decrease in peak calcium-pentagastrin stimulated calcitonin levels noted in 45 patients undergoing curative resection. The mean decrease achieved by these operations was 73.1%. In 22 (48%) of 45 patients, the postoperative stimulated calcitonin level dropped greater than 90% compared to the preoperative value.
Of the 45 patients who underwent reoperation with curative intent, 17 (38%) had postoperative stimulated calcitonin levels that were within the normal range (group 1), and 6 (13%) had no significant decrease in stimulated calcitonin levels (group 3). The remaining patients had a greater than 35% reduction in stimulated calcitonin levels (group 2). These three groups were compared regarding age, gender, disease type (sporadic or MEN-2A, or MEN-2B), age at diagnosis, number of previous operations, tumor size, and tumor invasiveness (defined as whether the primary tumor or its nodal metastases showed capsular invasion or invasion of adjacent structures). Tumor invasiveness was the only parameter that correlated with failure to reduce postoperative calcitonin levels to the normal range (p < 0.05, Fisher's exact test). In group 1 patients, review of disease from the primary operation did not show invasiveness in any case (0/17). In group 2 and 3 patients, invasiveness was identified in 8 of 28 cases.
Of the 17 patients who had normal postoperative stimulated calcitonin levels, 1-year follow-up levels are available for 12 patients (Table 3). In 10 of the 12 patients, the levels remained normal. Two patients (patient 2 and patient 5) had recurrent elevation of calcitonin levels, which have persisted after a 2-year follow-up. In five patients with normal calcitonin levels at 1 year, the calcitonin levels remained normal at 2-year follow-up.
In the seven patients who had palliative cervical operations, one patient had persistent postoperative hypocalcemia. There were no other complications in that group.
In the 45 patients who underwent reoperation with curative intent, there were no postoperative deaths and no transfusions were required. Thoracic duct leak occurred in four patients (8.9%). This required operative repair in each case. In two patients, the duct had been ligated at the time of surgery, but had leaked from a point proximal to the ligation. Operative ligation of the duct, through a cervical approach (three patients) or a thoracic approach (one patient) was successful in each case. Two patients have persistent hypocalcemia (follow-up, 3 months and 2 years). There were no permanent nerve injuries. Two patients had hoarseness that persisted for several weeks after surgery. Two patients had transient Horner's syndrome, and one patient had an axillary nerve praxis, which lasted for 1 month. One patient had a right hypoglossal nerve praxis with deviation of the tongue, which lasted for 2 months. Mean hospital stay was 4.2 ± 0.4 days (range, 1-14 days). All patients have resumed regular activity.
These results indicate an improvement in outcome after reoperation for persistent or recurrent MTC. In a previous series of patients, we reported initial normalization of calcitonin levels in 28% of patients and no change in levels in 31%.14 In the current series, postoperative stimulated calcitonin levels were in the normal range in 38%, and only 13% of patients did not have significantly reduced levels. Comparison of these results does not quite achieve statistical significance (p = 0.07, Fisher's exact test), but the trend is clear. This has occurred mainly through better preoperative selection of patients and the institution of routine direct examination of the liver before surgery, which identified metastases in ten patients, nine of whom had normal CT or MRI imaging of the liver and who would otherwise have undergone neck reoperation with curative intent. Despite obtaining CT or MRI of the neck in 32 patients who underwent cervical reoperation with curative intent, such imaging identified tumor in the neck that was not palpable in only five patients.
In two patients, no evidence of metastatic tumor was noted in the resected specimen (patients 5 and 9) (Table 3). Because of the decrease in calcitonin levels in both patients, it is likely that tumor was removed that was not detected by routine histologic examination. Patients 5 and 2 both showed a recurrence of hypercalcitoninemia 1 year after reoperation. Ten other patients still had a normal calcitonin level after 1 year, and five of these ten patients had normal levels after 2-year follow-up. The results are encouraging. It appears that based on this preliminary data, some patients who have normalization of calcitonin levels immediately after surgery will still have normal calcitonin levels after further follow-up. Given the indolent nature of this tumor, this must be considered preliminary data, and follow-up over the next decade is imperative to determine the long-term outcome from this operation.
The standard operation for primary medullary thyroid cancer is total thyroidectomy with central node clearance and ipsilateral functional or modified neck dissection for tumors >2 cm or if palpable adenopathy is present.2 Based on a review of the operative notes and pathology reports from previous operations and on the findings at reoperation, only 6 of the 45 patients undergoing neck dissection in this series had an adequate primary operation. In the other 39 cases, nodal tissue and, in some cases thyroid tissue, was found and removed from regions that should have been cleared by routine total thyroidectomy and node dissection. Given such information, one might question the need for cervical localization studies in most cases of persistent MTC. Careful dissection and systematic removal of the nodes that drain the thyroid will yield tumor in almost all cases. Conversely, detection of distant metastases continues to be of critical importance, and routine use of laparoscopy has been extremely useful in that regard.
In this series, the designation "normal" for stimulated calcitonin levels is based on the normal range for patients who have not been thyroidectomized. The results of stimulated calcitonin testing in patients who have been thyroidectomized for reasons other than MTC have not been well established. There is great variation between laboratories and techniques of measurement. Because all but three of the patients in this series are not from the St. Louis area, follow-up often is carried out by the patient's referring physician. In many cases, this physician is not trained in the performance of the calcium-pentagastrin test. In those cases, unstimulated levels, or levels measured after pentagastrin stimulation only, are obtained. This is not ideal, because calcium-pentagastrin stimulation is the most sensitive test; however, the sensitivity of the calcitonin assays now available (Nichols assay) may allow detection of significant unstimulated elevations if followed on a yearly basis.
In discussing reoperation with patients who have elevated calcitonin levels after primary surgery, the risks and options must be described in detail. We always discuss the option of continuing observation and cite recent series that have shown the frequently indolent nature of this tumor.10 Many patients are comfortable with a nonoperative expectant approach. In this series of 115 patients, 24 decided not to undergo further evaluation or surgical intervention for this problem. If a patient with elevated calcitonin levels has had an adequate previous operation and the results of imaging studies are negative, an expectant approach with routine yearly screening is appropriate in many cases. We do, however, think that it is important to follow these patients closely with routine CT or MRI scanning of the neck and chest. This surveillance should be carried out because if a central recurrence develops in the patients, it can be resected. This will lead to improved survival. Patients who die from MTC generally die from central recurrence or distant metastases. Central recurrence with airway invasion or vascular invasion or both is a catastrophic and preventable complication. It is our belief that even in the presence of distant metastases, local control by operative removal of tumor in the neck will result in improved survival and quality of life.
Reoperation for recurrent or residual MTC appears to be beneficial in the majority of patients and may result in improved local control rates and improved survival. These operations have been carried out with no mortality and minimum morbidity. Long-term follow-up is required to confirm the presumed benefit of these procedures.
The authors thank Dr. Samuel A. Wells and Dr. Lars Eric Tisell for guidance and Renea Daesch for assistance in typing the manuscript.
1. Wells S, Baylin S, Linehan W, et al. Provocative agents and the diagnosis of medullary carcinoma of the thyroid gland. Ann Surg 1978; 188:139-141. Library Holdings Bibliographic Links [Context Link]
2. Moley JF. Medullary thyroid cancer. Surg Clin North Am 1995; 75:405-420. Library Holdings Bibliographic Links [Context Link]
3. Miller HH, Melvin KEW, Gibson JM, Tashjian JH. Surgical approach to early familial medullary carcinoma of the thyroid gland. Am J Surg 1972; 123:438-443. Library Holdings Bibliographic Links [Context Link]
4. Block M, Jackson C, Greenwald K, et al. Clinical characteristics distinguishing hereditary from sporadic medullary thyroid carcinoma. Arch Surg 1980; 115:142-148. Library Holdings Bibliographic Links [Context Link]
5. Stepanas A, Samaan N, Hill C, Hickey R. Medullary thyroid carcinoma: importance of serial serum calcitonin measurement. Cancer 1979; 43:825-837. Library Holdings Bibliographic Links [Context Link]
6. Block M, Jackson C, Tashjian A. Management of occult medullary thyroid carcinoma: evidenced only by serum calcitonin elevations after apparently adequate neck operations. Arch Surg 1978; 113:368-372. Library Holdings Bibliographic Links [Context Link]
7. Dralle H, Damm I, Scheumann GF, et al. Compartment-oriented microdissection of regional lymph nodes in medullary thyroid carcinoma. Surg Today 1994; 24:112-121. Library Holdings Bibliographic Links [Context Link]
8. Norton J, Doppman J, Brennan M. Localization and resection of clinically inapparent medullary carcinoma of the thyroid. Surgery 1980; 87:616-622. Library Holdings Bibliographic Links [Context Link]
9. Tisell L, Hansson G, Jansson S, Salander H. Reoperation in the treatment of asymptomatic metastasizing medullary thyroid carcinoma. Surgery 1986; 99:60-66. Library Holdings Bibliographic Links [Context Link]
10. Van Heerden J, Grant C, Gharib H, et al. Long-term course of patients with persistent hypercalcitoninemia after apparent curative primary surgery for medullary thyroid carcinoma. Ann Surg 1990; 212:395-401. Library Holdings Bibliographic Links [Context Link]
11. Brumsen C, Haak H, Goslings B, Van de Velde C. Should patients with medullary thyroid carcinoma undergo extensive lymph node (re) operation to improve longterm survival? Henry Ford Hosp Med J 1992; 40:271-275. Library Holdings Bibliographic Links [Context Link]
12. Buhr H, Kallinowski F, Raue F, et al. Microsurgical neck dissection for occultly metastasizing medullary thyroid carcinoma: three year results. Cancer 1993; 72:3685-3693. Library Holdings Bibliographic Links [Context Link]
13. Ellenhorn JD, Shah JP, Brennan MF. Impact of therapeutic regional lymph node dissection for medullary carcinoma of the thyroid gland. Surgery 1993; 114:1078-1081. Library Holdings Bibliographic Links [Context Link]
14. Moley JF, Wells SA, Dilley WG, Tisell LE. Reoperation for recurrent or persistent medullary thyroid cancer. Surgery 1993; 114:1090-1095. Library Holdings Bibliographic Links [Context Link]
15. Tung W, Moley JF. Laparoscopic detection of hepatic metastases in patients with medullary thyroid cancer. Surgery 1995; 118:1024-1030. Library Holdings Bibliographic Links [Context Link]
16. Wells SA, Ross AJ, Dale JK, Gray RS. Transplantation of the parathyroid glands: current status. Surg Clin North Am 1979; 59:167-177. Library Holdings Bibliographic Links [Context Link]
Dr. Peter J. Fabri (Tampa, Florida): Thank you, Dr. Hanks, Dr. Copeland, Members, and Guests. First, Dr. Moley should be complimented on the outstanding technical results he has been able to accomplish in this group of patients.
In reviewing the manuscript, which he was kind enough to send to me, it is obvious that there is probably only one longterm complication in this entire group of reoperative patients, and that is hypocalcemia. All the rest of the complications were transient and treated by various methods. This is truly an outstanding result.
I am reminded of my first introduction to gastrinoma a number of years ago. At that time, we recognized that 75% were multiple, 50% were malignant, and 25% were metastatic. We also recognized that there were two diseases. There was one group of patients who had fulminant disease, and they died within a year or two, and another group that had very indolent disease that went on for 15, 20, or even 30 years.
Then serum gastrin radioimmunoassay was introduced, and we were able to diagnose the disease at a much earlier stage. And then calcium and finally secretin stimulation allowed us to clearly identify subgroups of patients. In our laboratory, we looked at the molecular fraction H-neu gastrin and were able to identify those individuals who were going to go on to develop aggressive disease and recurrence versus another group of people who had an almost 10-year survival.
But then the CAT scan came out and the MRI came out, and now in 1996, in almost any series that you look at, the 10-year survival is 100%. All the patients do well. It has become almost impossible to address the issue of outcome of our treatment. I am reminded that outcome is equal to an enumerator divided by a denominator, and Dr. Moley is addressing both the enumerator and the denominator.
He is applying a technique and looking to see whether that technique improves outcome, but he is also significantly affecting the number of patients that he treats. To that end I have a couple of questions.
First, in Dr. Moley's manuscript, he points out that invasiveness of the tumor was the only predictor of whether calcitonin levels would fall or not. I wonder, since this represents the biologic behavior of the tumor and not necessarily the technical operation, if he has looked at any other biologic markers or oncogenes to try to identify those patients who have the greatest likelihood of deriving benefit.
Second, Dr. Moley has chosen to use the normal calcitonin level in determining whether calcitonin returns to normal and not to a zero calcitonin level or the lowest level of detection in the laboratory. I am wondering how the results would be changed if he used nondetectable calcitonin as an indicator of a successful operation.
And, finally, there is a subgroup of 24 patients who, as far as I can tell, are identical to the other group, but who refused operative intervention. I wonder if you have had the opportunity to evaluate those patients and use them as a control group to see whether the changes in calcitonin over your two-year follow up are different in the operative patients from those in the nonoperative group.
It is going to be a long time, I suspect, until you can give us outcome data, but you should be congratulated for your outstanding work. I for one am quite pleased that it is possible to get such wonderful results in reoperative thyroid surgery. Thank you.
Dr. Roger S. Foster, Jr. (Atlanta, Georgia): Dr. Hanks, Dr. Copeland, Members, and Guests. I enjoyed the presentation of this paper and also the opportunity to read the full manuscript.
Even though medullary carcinomas of the thyroid represent 5% to 10% of thyroid cancers, it was not until the 1950s that Dr. Hazard, Dr. Hawk, and Dr. George Crile, Jr. first recognized that it was a distinct clinicopathologic entity. Since then, there has been great progress in understanding this disease, and this paper, I think, is a further contribution.
My endocrine surgery practice in Vermont was augmented by the opportunity to care for a large multiple endocrine neophasia (type II) kindred, four generations of whom were my patients. Based on my own experience and on review of the literature, some years ago I wrote that secondary resections for elevated calcitonin rarely have resulted in cures. The recent work of surgeons like Dr. Moley, Dr. Tisell, and others has apparently proven that statement wrong, at least in relation to neck and mediastinal nodal disease.
I suspect that long-term follow up of patients whose calcitonin levels have returned to normal will show that some, perhaps most, have remained free of disease. I look forward to data on the longer term follow up of both your operated patients and the patients who refused operation.
Unfortunately, the issue of the significance of elevated calcitonin levels after thyroidectomy for medullary carcinoma is complex, particularly in patients with occult disease. Some patients, perhaps many patients, who are clinically free of disease, yet have elevated calcitonins, remain clinically free of disease sometimes 3 to 4 decades after their total thyroidectomy. Such patients will be unlikely to benefit from reoperation.
How do we distinguish such patients from those where elevated calcitonin is a marker for disease that will become progressive? The only patients in this study who achieved normal calcitonin levels were those with less invasive tumors. Are they really the patients who need resection of occult disease, or are they the patients with indolent disease? I think follow up of the two groups will help.
I have four additional questions for Dr. Moley. You are apparently the only surgeon in your institution performing this type of reoperative surgery, despite the presence of several very experienced endocrine surgeons. My first question is, how different are your dissections from the nodal dissection performed at the primary operation by other skilled endocrine surgeons? Is special training needed?
Second, do you find any candidates for your reoperative surgery when the primary surgery has been done by surgeons skilled in the careful dissection of the cervical and mediastinal node metastases, or only after lesser dissections?
Third, you've described laparoscopy to exclude hepatic metastases. I am surprised that laparoscopy would be more sensitive than selective venous sampling, for example, from the superior vena cava. Would you please comment?
Finally, it is going to require long-term follow up to prove that the calcitonin levels remain normal. Do you have any information on the collective world experience following reoperative dissections of the type you have described?
Thank you.
Dr. Henry L. Laws (Birmingham, Alabama): Dr. Hanks, Dr. Copeland, Dr. Moley, and Associates. I would congratulate you on an awesome series of these reoperations.
I would like to ask you about one small thing. How did you handle the thoracic duct leaks? I assisted my associate, R. B. Kent, III, who had a persistent thoracic duct leak from reoperation for thyroid cancer. It was fairly simple to clip the thoracic duct within the chest thoracoscopically. That is really a very easy way to do it.
I just wonder if that is what you did, or do you have a better way? I enjoyed your paper very much. Thank you.
Dr. Murray F. Brennan (New York, New York): Thank you. The authors were kind enough to send me a copy of the manuscript, and I did review it.
I think Dr. Moley and his colleagues have a quite extraordinary experience. The data in the manuscript is even more extensive than the abstract, and I do not believe any individual in this country at the present time has the single-person experience as has had Dr. Moley.
That is a compliment to Dr. Wells and his colleagues because of their longstanding interest in medullary carcinoma of the thyroid that they are able to focus these endeavors in a single colleague. That is to be admired.
It is very clear that consistent elevation of thyrocalcitonin after operation for medullary carcinoma is common, and it is a difficult problem. The St. Louis group has taken a very aggressive approach to the entity. Of the 45 patients undergoing extensive neck resection, 16 were returned to peripheral stimulated normal calcitonin levels. The remaining 60% remained with elevated calcitonin levels and therefore were not biochemically cured, nor did they even have the option for biochemical cure. Then, of the 12 tested at one year, 2 had already increased their thyrocalcitonin levels.
Without minimizing the contribution in any way, I would take some minor issue with the authors that reoperation appears to be "curative" in a significant number of patients. As Dr. Moley alluded to, Dr. Jeff Norton, one of his colleagues, and myself, some years ago, following resection of subclinical disease identified by central vein catheterization, recatheterized those patients, and half of them, despite peripheral stimulated normal values, still had stimulatable central levels of thyrocalcitonin. They will certainly not be biochemically cured.
One, therefore, has to balance the yield in the 16 "cures" against the morbidity. As others have emphasized, most of the morbidity rate was transient; there were four thoracic duct leaks, two patients with permanent hypoparathyroidism, and several temporary nerve injuries. But, this is in the hands of a very experienced surgeon. We should support activities like this in very specialized units as the only hope for progress in difficult diseases, but in my opinion, this should not be standard practice.
I would ask the authors if they have recatheterized the 16 patients that are normal to see whether or not the levels in the neck remain elevated. I would also ask, as laparoscopy was performed rather than employing hepatic vein catheterization, but did do selective vein catheterization, is one modality for identifying metastatic disease better than another?
Is there information in this relatively short series of follow up, particularly as others have alluded to, in the 24 patients who declined operation?
I think the authors have to be complimented for quite extraordinary diligence. The chairman of the department is to be complimented on his ability to focus the endeavors in the hands of one surgeon. But, the results should not be translated outside of specialized units, and the results should be considered as a challenge for how to demonstrate that the individual patient has benefitted.
I thank you very much for asking me to read the manuscript, and I am sorry for keeping you so late. Thank you very much.
Dr. Jeffrey F. Moley (Closing Discussion): Thank you very much for your comments.
To respond to Dr. Fabri's question, in our 1993 series, we reported that the invasiveness of the primary tumor was also correlated with the inability to normalize calcitonin levels and, therefore, we did not discuss that today.
But it is clear that if the primary tumor invades local structures or if it demonstrates capsular invasion or, in a lymph node, if there is extracapsular extension, it is very unusual for us to be able to normalize the calcitonin levels.
That doesn't prevent me from operating on patients with invasive tumors if I feel that an inadequate primary operation was done. In this most recent series, I still was hopeful that some of them may be normalized; however, none of them were.
I also operate on these patients to prevent the long-term complication of central recurrence with tracheal invasion and vascular invasion, which is catastrophic and a horrible way for a patient to die and, I think, should be completely preventable by applying these kinds of operations to patients at an early stage.
As far as oncogenes are concerned, sporadic medullary thyroid cancer has been demonstrated to have mutations of the RET proto-oncogene in anywhere between 20% to 80% of cases, depending on which series you read. All hereditary MTC's are associated with germ-like RET mutations. In a series from Dr. Zedenius and Dr. Wallin from Sweden, they noted an association between codon 918 mutation in the RET proto-oncogene and aggressiveness of the tumor.
Aside from that, however, our group has found no correlation between other oncogenes such as p53. We have never found a p53 mutation in a medullary thyroid cancer (Herfarth et al., Genes, Chromosomes, and Cancer, in press.)
The 24 patients who refused the operation continue to do well, and their calcitonins continue to go up. I discuss the risks and the options in detail with these patients. I always discuss Dr. van Heerden's work, where he demonstrated an 86%, 15-year survival following primary surgery in patients who had persistent elevation of calcitonin levels.
Many patients decide not to undergo further evaluation or surgery. However, once their calcitonins continue to rise, they frequently request that we operate on them.
I think that the use of the word "cure" in patients with normalized calcitonins is a bit too strong. I think that the immediate normalization of levels in these patients is very encouraging, and I think it's a good start. I would call it a complete response in the terminology of the chemotherapist; however, we must wait long term to see if this has any long-term benefit. I'm pleased that we've been able to do the operations with low morbidity rates.
How do laparoscopy and selective venous catheterization compare in detecting liver metastases? Well, laparoscopy, in my opinion, is much better. We have gotten a number of false positives in patients with selective venous catheterization and elevated hepatic vein levels; we went ahead and did neck dissections in two of three patients, and the levels were normalized.
Those patients need to be followed, and if they do blossom with liver metastases in the future, SVC may be shown to be more accurate. But at this point, I've found the immediate usefulness of the laparoscopy to be much better than selective venous catheterization of the liver.
In the neck, selective venous catheterization is useful, but it only localizes tumor to a general area. Because we're going to be in these areas anyway, palpating the tissue and removing it, I have gotten away from doing routine selective venous catheterization except in selected patients, and work others up with CT, MRI, and laparoscopy.
As far as long-term follow-up by other groups in the world, really not much has been written. The German group led by Dr. Buhr reported a 3-year follow up in a series of over 50 patients. He does the Tisell procedure, and he reported normalization of calcitonin levels over a 3-year period in 25% of patients. I suspect that we're going to have similar results to that.
The other surgeons at Washington University are not doing this procedure, and I think there is probably an element of individual specialization there. However, they are certainly welcome to. I think that Dr. Norton, Dr. Doherty, Dr. Lairmore, and Dr. Wells are in agreement that this is a good approach with occasional reservations.
Have we recatheterized some of these patients with normalized calcitonin levels? No, we have not. As I said, only three of them reside in the St. Louis area, and most of them are from other states. Some are from Europe, Saudi Arabia, and Israel, and it's extremely difficult to do the kind of workup that I would like to do on a routine basis. It's hard to get a calcium pentagastrin stimulated calcitonin test on them.
To respond to Dr. Laws' question about thoracic duct leaks, we were able to explore the neck in three patients and do a direct cervical ligation. One patient, however, developed bilateral chylothorax, and in that patient we asked Dr. Alec Patterson to do a supradiaphragmatic mass ligation of the thoracic duct, and that was successful in stopping the leak.
I thank the Association for the privilege of presenting this data, and I thank the discussants for their questions.