World J Surg (2014) 38:645–652 DOI 10.1007/s00268-013-2379-9
Reoperative Experience with Papillary Thyroid Cancer Edwin O. Onkendi • Travis J. McKenzie • Melanie L. Richards David R. Farley • Geoffrey B. Thompson • Jan L. Kasperbauer Ian D. Hay • Clive S. Grant
• •
Published online: 4 December 2013 Ó Socie´te´ Internationale de Chirurgie 2013
Abstract Background Intense postoperative monitoring has resulted in increasing detection of patients with recurrent papillary thyroid cancer (PTC). Our goals included quantifying successful reoperation, and analyzing surgical complications and reasons for relapse. Methods From 1999 to 2008, a total of 410 patients underwent reoperation for PTC relapse. We analyzed postreoperative disease outcomes, reasons for relapse, and complications. Results Bilateral reoperative thyroidectomy was performed in 13 (3 %) patients; lobectomy, 34 (8 %); central neck (VI) soft tissue local recurrence excision, 47 (11.5 %); bilateral VI node dissection, 107 (26 %); unilateral VI dissection, 112 (27 %); levels II–V dissection, 93 (23 %); levels III–V, 86 (21 %); lateral single- or twocompartment dissection, 51 (12 %); and node picking, 20 (5 %) of level VI and 53 (13 %) lateral neck. Complications occurred in 6 %; including hypoparathyroidism, 3 %; unintentional recurrent laryngeal nerve (RLN) paralysis, 3 %; phrenic nerve injury, 0.5 %; spinal accessory nerve injury, 0.5 %; and chyle leak in 1.6 %. Of 380 (93 %) patients with follow-up (mean 5.2 years); 274 (72 %) patients are alive with no structural evidence of disease,
E. O. Onkendi T. J. McKenzie M. L. Richards D. R. Farley G. B. Thompson C. S. Grant (&) Department of Surgery, Mayo Clinic, Rochester, MN, USA e-mail: [email protected] J. L. Kasperbauer Department of Otolaryngology and Head and Neck Surgery, Mayo Clinic, Rochester, MN, USA I. D. Hay Division of Endocrinology, Mayo Clinic, Rochester, MN, USA
38 % developed disease relapse (mean 2.1 years), 42 (11 %) died from PTC, and 55 (14 %) are alive with disease. The reason for relapse was a false negative pre-reoperative ultrasound (US) in 18 (5 %), nodal recurrence in the operative field in 37 (10 %), a combination of these two reasons in 10 (3 %), and disease virulence (local or systemic recurrence) in 81 (21 %). Conclusions Although 72 % of patients were rendered structurally disease free after reoperation, nearly 40 % suffered additional relapse. Improved surgical technique or preoperative localization might positively affect 15–20 %; at least 20 % reflect the biologic aggressiveness of the disease.
Introduction Papillary thyroid cancer (PTC) accounts for nearly 85 % of thyroid cancers in the USA. The goal of initial operative treatment of PTC is to remove all malignant tumor tissue in the neck, including the primary tumor, tumor extending beyond the thyroid capsule, and all affected cervical lymph nodes while minimizing operative morbidity [1, 2]. Complete initial removal of disease, which is an important prognostic factor, facilitates subsequent accurate disease staging, postoperative surveillance with radioactive iodine (RAI) whole-body scanning (WBS) and monitoring of serum thyroglobulin (Tg), and any indicated treatment with RAI, either ablative or therapeutic, and minimizes risk of subsequent recurrence and metastasis [1]. While central neck dissection, either therapeutic or prophylactic, may be performed during thyroidectomy for PTC, prophylactic lateral neck dissection is not recommended. Therapeutic lateral neck dissection is performed for biopsyproven metastatic lateral cervical lymphadenopathy.
123
646
World J Surg (2014) 38:645–652
It is known that residual metastatic PTC in cervical lymph nodes is the most common type and site of disease persistence and subsequent relapse [3]. True soft tissue recurrence of PTC is uncommon but usually occurs in the operative field. Distant metastatic disease is also quite uncommon. Therefore, relapse of PTC may be classified as nodal, true soft tissue recurrences, or distant metastases. Relapse of PTC is increasingly being detected due to the intense postoperative monitoring that is carried out with RAI scanning, high-resolution ultrasound (US), and Tg monitoring—either suppressed or stimulated. Surgical excision is recommended as the first option for disease relapse [1], but is associated with potential risk of complications, including permanent hypoparathyroidism or injury to the recurrent laryngeal nerve, spinal accessory, or phrenic nerves, or thoracic duct with resultant chyle leak. The goals of our study were to quantify successful surgical resection, define reasons for further relapse of PTC, and characterize the surgical complications associated with reoperation for PTC at the Mayo Clinic (Rochester, MN, USA).
Materials and methods With the approval of the Mayo Clinic Institutional Review Board, we retrospectively reviewed the medical records of all patients who underwent their initial reoperation for recurrent or persistent PTC at the Mayo Clinic from January 1999 through December 2008. All patients whose records were reviewed in this study provided written consent to be involved in research. Some of the patients subsequently required additional operations and interventions (i.e. alcohol ablation) that are not specifically detailed, but incorporated in the final patient disease assessment. Patient demographics, laboratory findings, imaging studies, reoperative findings, surgical pathology, post-reoperative complications, and patient outcomes were reviewed. Operative notes and histopathology reports from the initial primary and subsequent reoperations performed prior to reoperations at our institution were reviewed when available. US scans were performed with high-resolution instruments (Sequoia; Acuson, Mountain View, CA, USA) equipped with high-frequency linear-array transducers of 7–13 MHz. US was used to derive a map that in one unified view identified the locations of the metastatic nodes within the four overall compartments (Fig. 1). Preoperative physical findings were assessed and correlated with preoperative US findings and these two were used to derive the following clinical findings: 1.
Palpable nodes, US additionally helpful. In patients with palpable nodes, US was considered helpful if pathologic non-palpable nodes were identified outside
123
Fig. 1 US map, prepared by the radiologist, confirmed ‘‘a 1 cm hypoechoic nodule in R mid-thyroid, previously biopsy-proven as PTC. No other thyroid masses.’’ It also showed ‘‘a 1.3 cm R midcervical LN suspicious for mets’’ (subsequently FNA positive for PTC). ‘‘Normal appearing L-sided, high R nodes.’’ The more linear appearing lymph nodes depicted in the high right and left IJ locations as well as the left neck imply they are normal-appearing. Subsequent pathology: ‘‘1.1 cm R tx pap ACA, 5 of 8 peri-tx nodes ? for PTC; L lobe normal. 1/10 mid, 2/10 low R jugular nodes ? for PTC.’’
2.
3.
the scope of what would have been the planned lymph node dissection indicated by the palpation findings. Therefore, even with clinically palpable disease, the US finding of pathologic non-palpable disease altered the operative plan. Non-palpable nodes, US helpful. The lymph node metastases (LNMs) were clinically non-palpable but US detected pathologic non-palpable disease and altered the operative plan. Palpable nodes, US not helpful. Palpable disease and the US findings that did not alter the operative plan.
In these reoperative patients, the accuracy of preoperative neck US was assessed separately for both the central and the lateral compartment nodes by correlating these findings with final histopathologic or cytologic results, or repeat US with corroborative evidence of disease [131iodine or positron emission tomography (PET) scan]. For purposes of US assessment, the right and left central neck (compartment VI; C-VI) and lateral neck (compartments II–V; C-II–V) were considered separately. The US findings were then classified into true positive, false positive, and false negative. True positive US accurately identified presence of all nodal or locally recurrent disease, which was verified histologically in the four potential areas: right, left C-VI; right, left lateral neck nodes (CII–V). False positive US predicted disease in one or more of the four areas, but no malignancy was found, either surgically or by subsequent repeated imaging.
World J Surg (2014) 38:645–652
False negative US predicted absence of disease but subsequent cytology or histopathology confirmed disease presence. Neck lymph node compartments were defined according to the 2008 classification by the American Head and Neck Society and the Committee for Head and Neck Surgery and Oncology of the American Academy of Otolaryngology– Head and Neck Surgery [4]. Disease relapse was based on the presence of structural disease, identified either by palpation or imaging; patients with elevations of serum Tg alone were not included. The types of disease relapses were subdivided into lymph node disease, true soft tissue local recurrence, and systemic recurrence. Nodal disease was defined pathologically as metastatic central or lateral neck lymph nodes. True local recurrence was defined as soft tissue disease without evidence of associated lymph node tissue. We assessed the causes of relapse and categorized them into three main reasons, or a combination of these. (1)
(2)
(3)
False negative US (radiologist responsible) If the prereoperative US failed to detect nodal disease in any one of the four areas, and the patient subsequently was demonstrated to have LNMs in that area, this was interpreted as a false-negative US. Recurrence in the operative field (surgeon responsible) If the surgeon operated in any of the four areas, and subsequent LNMs were discovered in the re-operated field, this was assessed as surgeon responsible. Disease virulence (virulent disease responsible) Any true local soft recurrence or development of distant metastases following reoperation was interpreted as due to disease virulence, not the fault of the surgeon or radiologist. If a patient had more than one cause of relapse and it included disease virulence, we considered this as outweighing the other reasons, and it was considered the primary cause.
Patient outcomes were classified as alive with no evidence of disease, alive with disease (AWD), dead with no disease, dead with disease (death due to a non-thyroid cancer etiology), and dead of disease (death due to thyroid cancer). Follow-up was calculated from the date of the first reoperation to the date of the relapse, and to the last known followup (first reoperation to last follow-up). Follow-up information was obtained from the medical record, correspondence from patients, and local physicians. Data are expressed as mean with percentages and range where appropriate.
647
recurrent/persistent PTC (details of the patients are tabulated in Table 1). There were 275 (67 %) females and 135 males (33 %), with mean age of 47 years (range 5–90 years). Of these, 325 (79 %) had undergone previous operations elsewhere, while 85 (21 %) patients had undergone their initial operation at our institution. The reoperations in this study were performed by five endocrine surgeons and four otolaryngology surgeons. Endocrine surgeons performed 315 (77 %) reoperations, while otolaryngology surgeons performed the remaining 95 (23 %) reoperations. Correlation of clinical and US findings Pre-reoperative US of the head and neck was obtained in 399 (97 %) patients, with sensitivity of 93 %, specificity of 80 %, positive predictive value (PPV) of 97 %, and accuracy of 92 %. A total of 126 (31 %) patients had palpable abnormal lymph nodes (60, central compartment; 66, lateral compartment), of whom 105 (25 %) underwent preoperative US. In this group, US was helpful in identifying additional pathologic non-palpable nodes, thereby changing the extent of the operation in 52 % of these patients. Of 342 patients who had no palpable pathologic central compartment lymph nodes, 142 (42 %) had non-palpable, pathologic central compartment nodes by US. In patients with no palpable abnormal lateral compartment lymph nodes, US identified pathologic nodes in 227 (67 %) of 339 patients. Pre-reoperation treatment Prior to our reoperations, 301 (73 %) patients received ablative radioactive 131Iodine with an average of 1.5 doses (range 1–6) and an average dose of 202 mCi (range 29.9–1,123). Patients had undergone an average of two prior operations (range 1–7), with an average of one central neck (range 1–6) and one lateral neck (range 1–5) operations prior to reoperation at our institution. Reoperations Thyroid and central neck local recurrences Bilateral reoperative thyroidectomy was performed in 13 (3 %) patients (total thyroidectomy, 7; near total thyroidectomy, 6) while completion thyroid lobectomy was performed in 36 (8 %) patients. As part of the central neck reoperation, 54 (13 %) patients had true soft tissue local recurrences resected.
Results
Central neck (C-VI) lymph nodes
During the study period from January 1999 to December 2008, a total of 410 consecutive patients underwent reoperation for
Central neck lymph node dissection was performed in 239 (58 %) patients. This encompassed a full bilateral central
123
648
World J Surg (2014) 38:645–652
Average, minimum/maximum
47, 5–90
Age [44 years
143
35
(24 %) patients. A selective node-picking procedure was carried out in 20 (8 %) patients. The average number of lymph nodes dissected from the central compartment was six, with 135 (57 %) patients having five or more lymph nodes removed. A total of 192 (80 %) patients had positive central compartment lymph nodes with the average number of positive lymph nodes of three.
142/227
40/67
Lateral neck (C III–V, ±II) lymph nodes
Table 1 Demographics, pre-reoperative features, reoperations, outcomes: 410 patients Characteristic
N
%
Gender, F/M
275/135
67/33
Age
Lymph nodes positive Non-palpable, US helpful: central/ lateral neck Palpable, overall, with US assessment
101
Palpable, US helpful
52
51
Palpable, US not helpful
49
49
Prior treatment with
131
Iodine
Patients
74 314
Average number of doses
1.5
Average dose (mCi), minimum/ maximum
202; 29.9/1,123
Prior neck operations Average/maximum central neck
1/6
Average/maximum lateral neck
1/5
Average/maximum total neck
2/7
Reoperative procedures Thyroid, local recurrence central neck
99
24
Bilateral thyroidectomy
13
3
Completion lobectomy
36
9
Excision local recurrence
54
13
239
69
Lymph node dissections Central neck Bilateral central neck
107
45
Central neck; right/left
54/58
23/24
Central neck node picking
20
8
Lateral neck Levels III–V Levels II–V
283
69
86 93
30 33
Dissection of one or two levels only
51
18
Lateral compartment node picking
53
19
US assessment True positive
290
73
False positive
8
2
False negative
22
6
273
72
Final disease status (380 pts with FU) Alive, no evidence of disease Alive, with disease
55
14
Dead of disease
42
11
Dead unrelated to disease
10
3
FU follow-up, US ultrasound
compartment lymph node dissection (left and right thyroid beds) in 107 (45 %) patients; the right central neck only was dissected in 54 (23 %) patients, and the left in 58
123
Lateral compartment lymph node dissection was performed in 283 (69 %) patients. This included compartments III–IV in 86 (30 % of the node dissections) patients, levels II–IV dissections in 93 (33 %) patients (including unilateral and bilateral), and selective single- or two-compartment dissections in 51 (18 %) patients. When a lateral neck dissection had already been performed, and the degree of scarring was confirmatory, a node picking procedure was performed in 53 (18 %) patients. The average number of lymph nodes removed during a lateral neck dissection was 17, with an average of four positive lymph nodes. Of the patients who had positive central or lateral compartment lymph nodes, 63 (16 %) had evidence of extranodal extension. Complications The unintentional complication rate following reoperation at our institution was 6 %. This was after excluding those patients who underwent intentional RLN sacrifice due to tumor involvement of the RLN and those who had prereoperative complications, including RLN injury and hypoparathyroidism. Hypoparathyroidism was present in 14 (3 %) patients prior to their reoperation. Hypoparathyroidism resulted from our reoperation in 13 (3 %) patients. Similarly, pre-existing recurrent laryngeal nerve paralysis was present in 28 (7 %) patients, whereas unintentional permanent recurrent laryngeal nerve paralysis following our procedures occurred in two patients. This did not include the ten (2 %) patients in whom the RLN was sacrificed due to involvement by disease. Other nerve injuries included phrenic nerve injury in 0.5 % of patients and spinal accessory nerve injury in 0.5 % of patients. Other complications included hematomas in 1.6 %, chyle leak in 1.7 %, and wound infection in 1.4 % (Table 2). Patient outcomes After reoperation, 380 (93 %) patients were followed for a mean of 5.2 years (range 0.1–13.5). Of these patients, 273 (72 %) patients were alive with no structural evidence of disease, 55 (14 %) were AWD, and 42 (11 %) patients had
World J Surg (2014) 38:645–652
649
Table 2 Post-reoperative complications Hypoparathyroidism Pre-reoperation, n (%)
14 (3)
Post-reoperation, n (%)
13 (3)
RLN Intentional sacrifice for disease, n (%)
10 (2)
Unintentional permanent RLN paralysis, n (%)
2 (0.5)
Phrenic nerve injury, n (%)
2 (0.5)
Spinal accessory nerve Temporary injury, n (%)
3 (0.7)
Permanent injury, n (%)
2 (0.5)
Cervical hematoma
7 (1.6)
Surgical site infection
6 (1.4)
Thoracic duct injury with chyle leak
7 (1.7)
RLN recurrent laryngeal nerve
Table 3 Reasons for relapse after reoperation Relapse (of 380 patients with FU), total (%) False-negative US, n (%)
146 (38) 18 (5)
Recurrence in the operative field, n (%)
37 (10)
Combined, false-negative US and recurrence in operative field, n (%)
10 (3)
Disease virulence, n (%)
81 (21)
Median time to relapse, years (range)
2.1 (0–8.6)
FU follow-up, US ultrasound
died of recurrent or persistent PTC (DOD). The latter two groups (AWD and DOD), constituting 25 % of all cases, represent a sub-selection of high-risk disease. Ten (3 %) patients died of non-PTC cause with no evidence of relapse. Post-reoperation recurrence and reasons for recurrence A total of 146 (38 %) patients who underwent reoperation developed subsequent relapse of PTC. Of these patients, 70 (48 %) had nodal recurrences only, whereas 76 (52 %) patients had a combination of local recurrence and/or systemic disease with or without additional nodal disease. The mean time to relapse following reoperation was 2.1 years. The reasons for relapse were a false-negative pre-reoperative US in 18 (5 %), nodal recurrence in the operative field in 37 (10 %), a combination of false-negative US and relapse in the operative field in ten (3 %) patients. Disease virulence was responsible for 21 % of all relapses (Table 3).
Discussion This contemporary, 10-year surgical series of 410 patients represents the integrated, multi-specialty effort at the
initial reoperation for which PTC relapse in the neck and mediastinum could potentially be resected. This incorporated resection of residual thyroid gland, central and lateral compartment lymph nodes, and soft tissue as indicated. The patients were widely heterogeneous, ranging over nine decades in age, nearly 75 % having been given an average of 200 mCi of RAI, and having undergone multiple previous neck operations to a maximum of seven involving both the central and the lateral neck. These patients all had structural disease, identifiable by either palpation and/or imaging. By this review, we intended to recognize where we had success in clearing disease and where we faced limitations that could potentially be overcome. But we also needed to understand the limitations imposed by the aggressive biology of the cancer in a subset of patients that defied our best efforts. Throughout this time period, and as a result of increasing experience, our surgical strategies evolved and technical nuances, particularly relating to safe preservation of important structures, were enhanced. It seemed logical to integrate our data and concepts evolution with results from the literature in a step-wise sequence of initial evaluation of the reoperative patients through their final outcome. Initial evaluation Whether by persistently elevated Tg levels or structural disease detected by palpation or imaging, the initial evaluation of the patient with PTC relapse optimally requires thorough evaluation, which includes the following: (a)
Prior operative and pathology notes; pathology slides These can be extremely helpful in appreciating the extent of thyroidectomy, lymph node dissection, key structures such as the RLNs and how many and which parathyroid glands were preserved. Details such as gross extra-thyroidal invasion or extra-nodal extension recognized by the surgeon or on histology may indicate more virulent disease. Total numbers of lymph nodes removed, the number metastatically involved, and from what specific compartments give an indication of initial disease location and burden. Was the disease totally excised or was gross disease residual present? To assess extent of the lateral neck dissection, was the omohyoid muscle dissected/sacrificed specifically? Was the thoracic duct seen/protected/ligated? Was the dissection extended to compartment II, and was the spinal accessory nerve identified? We invariably have outside slides evaluated by our pathologist, sometimes with additional findings such as a previously unrecognized parathyroid gland in the resected tissue.
123
650
World J Surg (2014) 38:645–652
(b)
intravenous contrast for optimum visualization, recognizing that this may delay the value and use of RAI. PET scan For patients who have evidence of more virulent disease, or where distant disease is suggested by the level of Tg or screening chest X-ray, PET/CT scan is utilized. This is most accurate when the RAI scan is negative and when more aggressive, perhaps higher-grade disease is present. The CT fused with the PET scan presently is of lower resolution than a highquality, dedicated CT, which may need to be obtained for more precise anatomic detail.
Physical examination Are metastatic nodes palpable? Fixed? Is the voice at all abnormal? Vocal cord function should always be visually assessed. (c) RAI The potential for administering RAI might be considered if the patient has not ever been treated or if the post-therapeutic scan continues to show uptake. (d) Reoperative timing We are reluctant to reoperate before about 3–4 months from the prior operation. The postoperative scarring improves in that interval, and recovery of any neurapraxia will often become apparent in that time. (e) Laboratory evaluation Serum Tg and Tg antibodies may be helpful in assessing burden of disease. Calcium ± parathyroid hormone (PTH) should be obtained—in our series, 3 % of patients had permanent hypoparathyroidism prior to our reoperation.
Imaging (a)
US central and lateral neck In contrast to imaging the central neck prior to initial thyroidectomy for PTC, where lymph nodes in compartment VI are not well delineated, in the reoperative setting when the thyroid gland has been resected, the right and left thyroid beds with the pretracheal and bilateral tracheoesophageal groove lymph nodes are quite accurately identified. In our series, 42 % of patients had nonpalpable pathologic nodes identified in the central neck, and 67 % in the lateral neck. Even when nodes were palpable, the addition of US changed the planned extent of the operation in over half of the patients. We consider US pivotal and critical in planning reoperation. Extremely helpful is a map that in one unified view identifies the locations of the metastatic nodes within the four overall compartments. US is considered a critical component to preoperative planning by Drs. Clayman and Tufano, experts in reoperative thyroid surgery [5–7]. We consider the right and left central and lateral neck compartments separately as four distinct overall compartments, and work carefully with the ultrasonographers to be as precise as possible in locating on this map the exact location of the abnormal nodes. (b) Computed tomography (CT) scan Although we have found US to be superior to CT scan in identifying metastatic lymph nodes in the central and lateral neck, CT is far superior in the assessment of central neck structures with invasive disease. Additionally, if nodes appear to extend inferiorly by US, we add CT to determine if disease continues into the mediastinum and to what extent. We do not hesitate to use
123
(c)
Operative strategy (a)
Central neck Only about 10 % of our patients required either unilateral or bilateral thyroid resection as part of the reoperative procedure. However, 58 % of our patients required central neck lymph node dissection. We could not ascertain from outside records whether the excised compartment VI nodes were macroscopically evident (i.e., not prophylactically excised) at the initial operation. Depending on either palpable or imaging indications, when technically feasible, we intended to clear the entire side of the central compartment as well as the pretracheal space. This often extended to the innominate artery inferiorly, but metastatic nodes were much less frequently found near the insertion of the RLN than inferiorly. This helped in protecting the area where the superior parathyroid gland would typically be found. Technically, the initial key step for safe dissection is to identify the carotid artery and clear its anterior surface from the thyroid cartilage to the base of the neck. Identification and dissection of the RLN was facilitated by dissecting low in the neck, below the level explored at the previous operation, often indicated by the presence of normal fat. Nerve monitoring was routinely used in about twothirds of the operations, but without any difference in RLN injury when not utilized. Magnification was often helpful. Autotransplantation of devascularized parathyroid glands was used, but most frequently parathyroid glands were not recognized during the dissection, yet the patients maintained normal calcium regulation postoperatively in all but 3 % of patients. We chose not to re-explore a central neck compartment if no indication by palpation or imaging of disease was present in order to avoid the risk of parathyroid devascularization. This is in contrast to the practice advised by Schuff et al. [8] and Steward [9]. (b) Lateral neck Lateral neck dissection was necessitated in 69 % of patients, with about equal proportion of compartments II–V, III–V, and selected single- or
World J Surg (2014) 38:645–652
(c)
two-compartment dissections. Limited dissections were used only if more extensive lateral neck dissections had previously occurred, either by our group, or as apparent when undertaking dissection in the operating room. Prior operative reports were not always accurate in describing the prior extent of dissection. If only compartment IV nodes were identified in a lateral compartment, we did not necessarily extend the superior dissection to encompass compartment II. We did not dissect a central or lateral neck compartment unless there was confirmed evidence of disease. The incisions were determined by surgeon preference, but often did require a vertical component depending on body habitus or ability to safely identify and expose the spinal accessory nerve. In contrast to the carotid sheath structures, phrenic nerve, and thoracic duct that are located on the periphery of the nodal packet, the spinal accessory and cervical plexus nerves run through this packet and need careful dissection to preserve them. Dissection posterior to the carotid artery is rarely necessary and jeopardizes the cervical sympathetic chain. Potentially overlooked are nodes anterior to the internal jugular vein and bifurcation of the carotid artery. Nodes low in the internal jugular chain— compartment IV—are common in PTC, which requires careful dissection to either preserve the thoracic duct, or ensure it is ligated, sometimes with several contributing branches. Operative complications Not surprisingly, the 17 % complication rate in this selected patient series was markedly higher than in first-time thyroid cancer operations. In patients having undergone multiple prior operations for more virulent disease, the more significant complications of hypoparathyroidism and nerve damage were comparable to other reported series [6, 10], but higher than reports of no complications when disease was much more modest [11] or the dissections were limited to the central neck [12].
651
structurally persistent disease (14 %). In fact, nearly 40 % of the reoperative patients suffered relapse subsequent to our reoperation at a mean of 2.1 years, nearly equally split between further nodal recurrence only, or some combination of true soft tissue recurrence and/or systemic disease with or without additional nodal disease. Recognizing that exact characterization between local recurrence and lymph node disease is not perfect, the patients with subsequent lymph node disease relapse we felt deserved careful scrutiny. They have disease that should potentially pose limited threat to life and perhaps be amenable to improved local control. Holding radiologists accountable for any lymph node disease discovered in any compartment at any time subsequent to the reoperation that was not identified on the initial preoperative scan is probably too strict, but there is potential to further reduce relapse in 5–8 % of patients. Even more patients (10 %) relapsed with additional lymph nodes in the previously operated field—more opportunity for improvement. Not analyzed were any subsequent reoperations that were necessary in a proportion of these patients. Even with relapse limited only to lymph node disease, the potential risk of damage to important structures associated with additional reoperation prompted a new avenue of thinking and treatment—US-guided alcohol ablation. The risks, efficacy, and cost analysis of this exciting adjunct intervention have been reported separately [14]. Soft tissue recurrence and/or systemic relapse represents the kind of cancer recurrence more commonly associated with higher-grade malignancies such as breast or colon carcinomas. To identify these higher-risk patients with PTC previously garnered considerable research attention and led to the development of important scoring systems [15]. The prognostic factors making up these scoring systems, such as aggressive histology, large tumor size, age over 45 years, and extra-thyroid extension have also been associated with reduced survival in the reoperative setting [16]. However, death due to PTC was distinctly rare in the usually reported, initially operated series of patients, quite different from this reoperative patient cohort, where such patients are over-represented.
Patient outcomes Conclusions We were gratified that nearly three-quarters of our patients at last follow-up had no structural evidence of disease. However, this is in stark contrast to disease recurrence being prevented in 95 % of patients after initial operation in our experience when optimal principles were followed of preoperative neck US, bilateral thyroidectomy, routine central neck dissection, and lateral neck dissection for disease [13]. Strikingly, 25 % of the reoperative patients had either died of disease (11 %) or were alive with
Although 72 % of reoperative patients can be rendered structurally disease-free after reoperation, nearly 40 % suffer additional relapse. Alterations in surgical technique or improved preoperative localization might affect about 15–20 % of patients; at least 20 % of the patients reflect increased biologic aggressiveness of the disease. The rate of complications, although greater than initial operations, seems acceptable.
123
652
References 1. Cooper D, Doherty G, Haugen B et al (2009) Revised American Thyroid Association management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid 19(11):1167–1214 2. Schlumberger M (1998) Papillary and follicular thyroid carcinoma. N Engl J Med 338(5):297–306 3. Wang T, Dubner S, Sznyter L et al (2004) Incidence of metastatic well-differentiated thyroid cancer in cervical lymph nodes. Arch Otolaryngol Head Neck Surg 130:110–113 4. Robbins K, Shaha A, Medina J et al (2008) Consensus statement on the classification and terminology of neck dissection. Arch Otolaryngol Head Neck Surg 134:536–538 5. Clayman G, Shellenberger T, Ginsberg L et al (2009) Approach and safety of comprehensive central compartment dissection in patients with recurrent papillary thyroid carcinoma. Head Neck 31(9):1152–1163 6. Tufano R, Bishop J, Wu G (2012) Reoperative central compartment dissection for patients with recurrent/persistent papillary thyroid cancer: efficacy, safety, and the association of the BRAF mutation. Laryngoscope 122:1634–1640 7. Farrag TY, Agrawal N, Sheth S et al (2007) Algorithm for safe and effective reoperative thyroid bed surgery for recurrent/persistent papillary thyroid carcinoma. Head Neck 29(12):1069–1074 8. Schuff K, Weber S, Givi B et al (2008) Efficacy of nodal dissection for treatment of persistent/recurrent papillary thyroid cancer. Laryngoscope 118:768–775 9. Steward D (2012) Update in utility of secondary node dissection for papillary thyroid cancer. J Clin Endocrin Metab 97(10):3393–3398
123
World J Surg (2014) 38:645–652 10. Shah M, Harris K, Nassif R et al (2012) Efficacy and safety of central compartment neck dissection for recurrent thyroid carcinoma. Arch Otolaryngol Head Neck Surg 138(1):33–37 11. Al-Saif O, Farrar W, Bloomston M et al (2010) Long-term efficacy of lymph node reoperation for persistent papillary thyroid cancer. J Clin Endocrin Metab 95:2187–2194 12. Alvarado R, Sywak M, Delbridge L et al (2009) Central lymph node dissection as a secondary procedure for papillary thyroid cancer: is there added morbidity? Surgery 145:514–518 13. Grant C, Stulak J, Thompson G et al (2010) Risks and adequacy of an optimized surgical approach to the primary surgical management of papillary thyroid carcinoma treated during 1999–2006. World J Surg 34:1239–1246. doi:10.1007/ s00268-009-0307-9 14. Hay ID, Lee RA, Davidge-Pitts C et al (2013) Long-term outcome of ultrasound-guided percutaneous ethanol ablation of selected ‘‘recurrent’’ neck nodal metastases in 25 patients with TNM stages III or IVA papillary thyroid carcinoma previously treated by surgery and (131)I therapy. Surgery 154(6):1448–1455 15. Hay ID, Bergstralh EJ, Goellner JR et al (1993) Predicting outcome in papillary thyroid carcinoma: development of a reliable prognostic scoring system in a cohort of 1779 patients surgically treated at one institution during 1940 through 1989. Surgery 114:1050–1058 16. Ito Y, Higashiyama T, Takamura Y et al (2011) Prognosis of patients with papillary thyroid carcinoma showing postoperative recurrence to the central neck. World J Surg 35:767–772. doi:10. 1007/s00268-010-0924-3
Reoperative Experience with Papillary Thyroid Cancer Edwin O. Onkendi • Travis J. McKenzie • Melanie L. Richards David R. Farley • Geoffrey B. Thompson • Jan L. Kasperbauer Ian D. Hay • Clive S. Grant
• •
Published online: 4 December 2013 Ó Socie´te´ Internationale de Chirurgie 2013
Abstract Background Intense postoperative monitoring has resulted in increasing detection of patients with recurrent papillary thyroid cancer (PTC). Our goals included quantifying successful reoperation, and analyzing surgical complications and reasons for relapse. Methods From 1999 to 2008, a total of 410 patients underwent reoperation for PTC relapse. We analyzed postreoperative disease outcomes, reasons for relapse, and complications. Results Bilateral reoperative thyroidectomy was performed in 13 (3 %) patients; lobectomy, 34 (8 %); central neck (VI) soft tissue local recurrence excision, 47 (11.5 %); bilateral VI node dissection, 107 (26 %); unilateral VI dissection, 112 (27 %); levels II–V dissection, 93 (23 %); levels III–V, 86 (21 %); lateral single- or twocompartment dissection, 51 (12 %); and node picking, 20 (5 %) of level VI and 53 (13 %) lateral neck. Complications occurred in 6 %; including hypoparathyroidism, 3 %; unintentional recurrent laryngeal nerve (RLN) paralysis, 3 %; phrenic nerve injury, 0.5 %; spinal accessory nerve injury, 0.5 %; and chyle leak in 1.6 %. Of 380 (93 %) patients with follow-up (mean 5.2 years); 274 (72 %) patients are alive with no structural evidence of disease,
E. O. Onkendi T. J. McKenzie M. L. Richards D. R. Farley G. B. Thompson C. S. Grant (&) Department of Surgery, Mayo Clinic, Rochester, MN, USA e-mail: [email protected] J. L. Kasperbauer Department of Otolaryngology and Head and Neck Surgery, Mayo Clinic, Rochester, MN, USA I. D. Hay Division of Endocrinology, Mayo Clinic, Rochester, MN, USA
38 % developed disease relapse (mean 2.1 years), 42 (11 %) died from PTC, and 55 (14 %) are alive with disease. The reason for relapse was a false negative pre-reoperative ultrasound (US) in 18 (5 %), nodal recurrence in the operative field in 37 (10 %), a combination of these two reasons in 10 (3 %), and disease virulence (local or systemic recurrence) in 81 (21 %). Conclusions Although 72 % of patients were rendered structurally disease free after reoperation, nearly 40 % suffered additional relapse. Improved surgical technique or preoperative localization might positively affect 15–20 %; at least 20 % reflect the biologic aggressiveness of the disease.
Introduction Papillary thyroid cancer (PTC) accounts for nearly 85 % of thyroid cancers in the USA. The goal of initial operative treatment of PTC is to remove all malignant tumor tissue in the neck, including the primary tumor, tumor extending beyond the thyroid capsule, and all affected cervical lymph nodes while minimizing operative morbidity [1, 2]. Complete initial removal of disease, which is an important prognostic factor, facilitates subsequent accurate disease staging, postoperative surveillance with radioactive iodine (RAI) whole-body scanning (WBS) and monitoring of serum thyroglobulin (Tg), and any indicated treatment with RAI, either ablative or therapeutic, and minimizes risk of subsequent recurrence and metastasis [1]. While central neck dissection, either therapeutic or prophylactic, may be performed during thyroidectomy for PTC, prophylactic lateral neck dissection is not recommended. Therapeutic lateral neck dissection is performed for biopsyproven metastatic lateral cervical lymphadenopathy.
123
646
World J Surg (2014) 38:645–652
It is known that residual metastatic PTC in cervical lymph nodes is the most common type and site of disease persistence and subsequent relapse [3]. True soft tissue recurrence of PTC is uncommon but usually occurs in the operative field. Distant metastatic disease is also quite uncommon. Therefore, relapse of PTC may be classified as nodal, true soft tissue recurrences, or distant metastases. Relapse of PTC is increasingly being detected due to the intense postoperative monitoring that is carried out with RAI scanning, high-resolution ultrasound (US), and Tg monitoring—either suppressed or stimulated. Surgical excision is recommended as the first option for disease relapse [1], but is associated with potential risk of complications, including permanent hypoparathyroidism or injury to the recurrent laryngeal nerve, spinal accessory, or phrenic nerves, or thoracic duct with resultant chyle leak. The goals of our study were to quantify successful surgical resection, define reasons for further relapse of PTC, and characterize the surgical complications associated with reoperation for PTC at the Mayo Clinic (Rochester, MN, USA).
Materials and methods With the approval of the Mayo Clinic Institutional Review Board, we retrospectively reviewed the medical records of all patients who underwent their initial reoperation for recurrent or persistent PTC at the Mayo Clinic from January 1999 through December 2008. All patients whose records were reviewed in this study provided written consent to be involved in research. Some of the patients subsequently required additional operations and interventions (i.e. alcohol ablation) that are not specifically detailed, but incorporated in the final patient disease assessment. Patient demographics, laboratory findings, imaging studies, reoperative findings, surgical pathology, post-reoperative complications, and patient outcomes were reviewed. Operative notes and histopathology reports from the initial primary and subsequent reoperations performed prior to reoperations at our institution were reviewed when available. US scans were performed with high-resolution instruments (Sequoia; Acuson, Mountain View, CA, USA) equipped with high-frequency linear-array transducers of 7–13 MHz. US was used to derive a map that in one unified view identified the locations of the metastatic nodes within the four overall compartments (Fig. 1). Preoperative physical findings were assessed and correlated with preoperative US findings and these two were used to derive the following clinical findings: 1.
Palpable nodes, US additionally helpful. In patients with palpable nodes, US was considered helpful if pathologic non-palpable nodes were identified outside
123
Fig. 1 US map, prepared by the radiologist, confirmed ‘‘a 1 cm hypoechoic nodule in R mid-thyroid, previously biopsy-proven as PTC. No other thyroid masses.’’ It also showed ‘‘a 1.3 cm R midcervical LN suspicious for mets’’ (subsequently FNA positive for PTC). ‘‘Normal appearing L-sided, high R nodes.’’ The more linear appearing lymph nodes depicted in the high right and left IJ locations as well as the left neck imply they are normal-appearing. Subsequent pathology: ‘‘1.1 cm R tx pap ACA, 5 of 8 peri-tx nodes ? for PTC; L lobe normal. 1/10 mid, 2/10 low R jugular nodes ? for PTC.’’
2.
3.
the scope of what would have been the planned lymph node dissection indicated by the palpation findings. Therefore, even with clinically palpable disease, the US finding of pathologic non-palpable disease altered the operative plan. Non-palpable nodes, US helpful. The lymph node metastases (LNMs) were clinically non-palpable but US detected pathologic non-palpable disease and altered the operative plan. Palpable nodes, US not helpful. Palpable disease and the US findings that did not alter the operative plan.
In these reoperative patients, the accuracy of preoperative neck US was assessed separately for both the central and the lateral compartment nodes by correlating these findings with final histopathologic or cytologic results, or repeat US with corroborative evidence of disease [131iodine or positron emission tomography (PET) scan]. For purposes of US assessment, the right and left central neck (compartment VI; C-VI) and lateral neck (compartments II–V; C-II–V) were considered separately. The US findings were then classified into true positive, false positive, and false negative. True positive US accurately identified presence of all nodal or locally recurrent disease, which was verified histologically in the four potential areas: right, left C-VI; right, left lateral neck nodes (CII–V). False positive US predicted disease in one or more of the four areas, but no malignancy was found, either surgically or by subsequent repeated imaging.
World J Surg (2014) 38:645–652
False negative US predicted absence of disease but subsequent cytology or histopathology confirmed disease presence. Neck lymph node compartments were defined according to the 2008 classification by the American Head and Neck Society and the Committee for Head and Neck Surgery and Oncology of the American Academy of Otolaryngology– Head and Neck Surgery [4]. Disease relapse was based on the presence of structural disease, identified either by palpation or imaging; patients with elevations of serum Tg alone were not included. The types of disease relapses were subdivided into lymph node disease, true soft tissue local recurrence, and systemic recurrence. Nodal disease was defined pathologically as metastatic central or lateral neck lymph nodes. True local recurrence was defined as soft tissue disease without evidence of associated lymph node tissue. We assessed the causes of relapse and categorized them into three main reasons, or a combination of these. (1)
(2)
(3)
False negative US (radiologist responsible) If the prereoperative US failed to detect nodal disease in any one of the four areas, and the patient subsequently was demonstrated to have LNMs in that area, this was interpreted as a false-negative US. Recurrence in the operative field (surgeon responsible) If the surgeon operated in any of the four areas, and subsequent LNMs were discovered in the re-operated field, this was assessed as surgeon responsible. Disease virulence (virulent disease responsible) Any true local soft recurrence or development of distant metastases following reoperation was interpreted as due to disease virulence, not the fault of the surgeon or radiologist. If a patient had more than one cause of relapse and it included disease virulence, we considered this as outweighing the other reasons, and it was considered the primary cause.
Patient outcomes were classified as alive with no evidence of disease, alive with disease (AWD), dead with no disease, dead with disease (death due to a non-thyroid cancer etiology), and dead of disease (death due to thyroid cancer). Follow-up was calculated from the date of the first reoperation to the date of the relapse, and to the last known followup (first reoperation to last follow-up). Follow-up information was obtained from the medical record, correspondence from patients, and local physicians. Data are expressed as mean with percentages and range where appropriate.
647
recurrent/persistent PTC (details of the patients are tabulated in Table 1). There were 275 (67 %) females and 135 males (33 %), with mean age of 47 years (range 5–90 years). Of these, 325 (79 %) had undergone previous operations elsewhere, while 85 (21 %) patients had undergone their initial operation at our institution. The reoperations in this study were performed by five endocrine surgeons and four otolaryngology surgeons. Endocrine surgeons performed 315 (77 %) reoperations, while otolaryngology surgeons performed the remaining 95 (23 %) reoperations. Correlation of clinical and US findings Pre-reoperative US of the head and neck was obtained in 399 (97 %) patients, with sensitivity of 93 %, specificity of 80 %, positive predictive value (PPV) of 97 %, and accuracy of 92 %. A total of 126 (31 %) patients had palpable abnormal lymph nodes (60, central compartment; 66, lateral compartment), of whom 105 (25 %) underwent preoperative US. In this group, US was helpful in identifying additional pathologic non-palpable nodes, thereby changing the extent of the operation in 52 % of these patients. Of 342 patients who had no palpable pathologic central compartment lymph nodes, 142 (42 %) had non-palpable, pathologic central compartment nodes by US. In patients with no palpable abnormal lateral compartment lymph nodes, US identified pathologic nodes in 227 (67 %) of 339 patients. Pre-reoperation treatment Prior to our reoperations, 301 (73 %) patients received ablative radioactive 131Iodine with an average of 1.5 doses (range 1–6) and an average dose of 202 mCi (range 29.9–1,123). Patients had undergone an average of two prior operations (range 1–7), with an average of one central neck (range 1–6) and one lateral neck (range 1–5) operations prior to reoperation at our institution. Reoperations Thyroid and central neck local recurrences Bilateral reoperative thyroidectomy was performed in 13 (3 %) patients (total thyroidectomy, 7; near total thyroidectomy, 6) while completion thyroid lobectomy was performed in 36 (8 %) patients. As part of the central neck reoperation, 54 (13 %) patients had true soft tissue local recurrences resected.
Results
Central neck (C-VI) lymph nodes
During the study period from January 1999 to December 2008, a total of 410 consecutive patients underwent reoperation for
Central neck lymph node dissection was performed in 239 (58 %) patients. This encompassed a full bilateral central
123
648
World J Surg (2014) 38:645–652
Average, minimum/maximum
47, 5–90
Age [44 years
143
35
(24 %) patients. A selective node-picking procedure was carried out in 20 (8 %) patients. The average number of lymph nodes dissected from the central compartment was six, with 135 (57 %) patients having five or more lymph nodes removed. A total of 192 (80 %) patients had positive central compartment lymph nodes with the average number of positive lymph nodes of three.
142/227
40/67
Lateral neck (C III–V, ±II) lymph nodes
Table 1 Demographics, pre-reoperative features, reoperations, outcomes: 410 patients Characteristic
N
%
Gender, F/M
275/135
67/33
Age
Lymph nodes positive Non-palpable, US helpful: central/ lateral neck Palpable, overall, with US assessment
101
Palpable, US helpful
52
51
Palpable, US not helpful
49
49
Prior treatment with
131
Iodine
Patients
74 314
Average number of doses
1.5
Average dose (mCi), minimum/ maximum
202; 29.9/1,123
Prior neck operations Average/maximum central neck
1/6
Average/maximum lateral neck
1/5
Average/maximum total neck
2/7
Reoperative procedures Thyroid, local recurrence central neck
99
24
Bilateral thyroidectomy
13
3
Completion lobectomy
36
9
Excision local recurrence
54
13
239
69
Lymph node dissections Central neck Bilateral central neck
107
45
Central neck; right/left
54/58
23/24
Central neck node picking
20
8
Lateral neck Levels III–V Levels II–V
283
69
86 93
30 33
Dissection of one or two levels only
51
18
Lateral compartment node picking
53
19
US assessment True positive
290
73
False positive
8
2
False negative
22
6
273
72
Final disease status (380 pts with FU) Alive, no evidence of disease Alive, with disease
55
14
Dead of disease
42
11
Dead unrelated to disease
10
3
FU follow-up, US ultrasound
compartment lymph node dissection (left and right thyroid beds) in 107 (45 %) patients; the right central neck only was dissected in 54 (23 %) patients, and the left in 58
123
Lateral compartment lymph node dissection was performed in 283 (69 %) patients. This included compartments III–IV in 86 (30 % of the node dissections) patients, levels II–IV dissections in 93 (33 %) patients (including unilateral and bilateral), and selective single- or two-compartment dissections in 51 (18 %) patients. When a lateral neck dissection had already been performed, and the degree of scarring was confirmatory, a node picking procedure was performed in 53 (18 %) patients. The average number of lymph nodes removed during a lateral neck dissection was 17, with an average of four positive lymph nodes. Of the patients who had positive central or lateral compartment lymph nodes, 63 (16 %) had evidence of extranodal extension. Complications The unintentional complication rate following reoperation at our institution was 6 %. This was after excluding those patients who underwent intentional RLN sacrifice due to tumor involvement of the RLN and those who had prereoperative complications, including RLN injury and hypoparathyroidism. Hypoparathyroidism was present in 14 (3 %) patients prior to their reoperation. Hypoparathyroidism resulted from our reoperation in 13 (3 %) patients. Similarly, pre-existing recurrent laryngeal nerve paralysis was present in 28 (7 %) patients, whereas unintentional permanent recurrent laryngeal nerve paralysis following our procedures occurred in two patients. This did not include the ten (2 %) patients in whom the RLN was sacrificed due to involvement by disease. Other nerve injuries included phrenic nerve injury in 0.5 % of patients and spinal accessory nerve injury in 0.5 % of patients. Other complications included hematomas in 1.6 %, chyle leak in 1.7 %, and wound infection in 1.4 % (Table 2). Patient outcomes After reoperation, 380 (93 %) patients were followed for a mean of 5.2 years (range 0.1–13.5). Of these patients, 273 (72 %) patients were alive with no structural evidence of disease, 55 (14 %) were AWD, and 42 (11 %) patients had
World J Surg (2014) 38:645–652
649
Table 2 Post-reoperative complications Hypoparathyroidism Pre-reoperation, n (%)
14 (3)
Post-reoperation, n (%)
13 (3)
RLN Intentional sacrifice for disease, n (%)
10 (2)
Unintentional permanent RLN paralysis, n (%)
2 (0.5)
Phrenic nerve injury, n (%)
2 (0.5)
Spinal accessory nerve Temporary injury, n (%)
3 (0.7)
Permanent injury, n (%)
2 (0.5)
Cervical hematoma
7 (1.6)
Surgical site infection
6 (1.4)
Thoracic duct injury with chyle leak
7 (1.7)
RLN recurrent laryngeal nerve
Table 3 Reasons for relapse after reoperation Relapse (of 380 patients with FU), total (%) False-negative US, n (%)
146 (38) 18 (5)
Recurrence in the operative field, n (%)
37 (10)
Combined, false-negative US and recurrence in operative field, n (%)
10 (3)
Disease virulence, n (%)
81 (21)
Median time to relapse, years (range)
2.1 (0–8.6)
FU follow-up, US ultrasound
died of recurrent or persistent PTC (DOD). The latter two groups (AWD and DOD), constituting 25 % of all cases, represent a sub-selection of high-risk disease. Ten (3 %) patients died of non-PTC cause with no evidence of relapse. Post-reoperation recurrence and reasons for recurrence A total of 146 (38 %) patients who underwent reoperation developed subsequent relapse of PTC. Of these patients, 70 (48 %) had nodal recurrences only, whereas 76 (52 %) patients had a combination of local recurrence and/or systemic disease with or without additional nodal disease. The mean time to relapse following reoperation was 2.1 years. The reasons for relapse were a false-negative pre-reoperative US in 18 (5 %), nodal recurrence in the operative field in 37 (10 %), a combination of false-negative US and relapse in the operative field in ten (3 %) patients. Disease virulence was responsible for 21 % of all relapses (Table 3).
Discussion This contemporary, 10-year surgical series of 410 patients represents the integrated, multi-specialty effort at the
initial reoperation for which PTC relapse in the neck and mediastinum could potentially be resected. This incorporated resection of residual thyroid gland, central and lateral compartment lymph nodes, and soft tissue as indicated. The patients were widely heterogeneous, ranging over nine decades in age, nearly 75 % having been given an average of 200 mCi of RAI, and having undergone multiple previous neck operations to a maximum of seven involving both the central and the lateral neck. These patients all had structural disease, identifiable by either palpation and/or imaging. By this review, we intended to recognize where we had success in clearing disease and where we faced limitations that could potentially be overcome. But we also needed to understand the limitations imposed by the aggressive biology of the cancer in a subset of patients that defied our best efforts. Throughout this time period, and as a result of increasing experience, our surgical strategies evolved and technical nuances, particularly relating to safe preservation of important structures, were enhanced. It seemed logical to integrate our data and concepts evolution with results from the literature in a step-wise sequence of initial evaluation of the reoperative patients through their final outcome. Initial evaluation Whether by persistently elevated Tg levels or structural disease detected by palpation or imaging, the initial evaluation of the patient with PTC relapse optimally requires thorough evaluation, which includes the following: (a)
Prior operative and pathology notes; pathology slides These can be extremely helpful in appreciating the extent of thyroidectomy, lymph node dissection, key structures such as the RLNs and how many and which parathyroid glands were preserved. Details such as gross extra-thyroidal invasion or extra-nodal extension recognized by the surgeon or on histology may indicate more virulent disease. Total numbers of lymph nodes removed, the number metastatically involved, and from what specific compartments give an indication of initial disease location and burden. Was the disease totally excised or was gross disease residual present? To assess extent of the lateral neck dissection, was the omohyoid muscle dissected/sacrificed specifically? Was the thoracic duct seen/protected/ligated? Was the dissection extended to compartment II, and was the spinal accessory nerve identified? We invariably have outside slides evaluated by our pathologist, sometimes with additional findings such as a previously unrecognized parathyroid gland in the resected tissue.
123
650
World J Surg (2014) 38:645–652
(b)
intravenous contrast for optimum visualization, recognizing that this may delay the value and use of RAI. PET scan For patients who have evidence of more virulent disease, or where distant disease is suggested by the level of Tg or screening chest X-ray, PET/CT scan is utilized. This is most accurate when the RAI scan is negative and when more aggressive, perhaps higher-grade disease is present. The CT fused with the PET scan presently is of lower resolution than a highquality, dedicated CT, which may need to be obtained for more precise anatomic detail.
Physical examination Are metastatic nodes palpable? Fixed? Is the voice at all abnormal? Vocal cord function should always be visually assessed. (c) RAI The potential for administering RAI might be considered if the patient has not ever been treated or if the post-therapeutic scan continues to show uptake. (d) Reoperative timing We are reluctant to reoperate before about 3–4 months from the prior operation. The postoperative scarring improves in that interval, and recovery of any neurapraxia will often become apparent in that time. (e) Laboratory evaluation Serum Tg and Tg antibodies may be helpful in assessing burden of disease. Calcium ± parathyroid hormone (PTH) should be obtained—in our series, 3 % of patients had permanent hypoparathyroidism prior to our reoperation.
Imaging (a)
US central and lateral neck In contrast to imaging the central neck prior to initial thyroidectomy for PTC, where lymph nodes in compartment VI are not well delineated, in the reoperative setting when the thyroid gland has been resected, the right and left thyroid beds with the pretracheal and bilateral tracheoesophageal groove lymph nodes are quite accurately identified. In our series, 42 % of patients had nonpalpable pathologic nodes identified in the central neck, and 67 % in the lateral neck. Even when nodes were palpable, the addition of US changed the planned extent of the operation in over half of the patients. We consider US pivotal and critical in planning reoperation. Extremely helpful is a map that in one unified view identifies the locations of the metastatic nodes within the four overall compartments. US is considered a critical component to preoperative planning by Drs. Clayman and Tufano, experts in reoperative thyroid surgery [5–7]. We consider the right and left central and lateral neck compartments separately as four distinct overall compartments, and work carefully with the ultrasonographers to be as precise as possible in locating on this map the exact location of the abnormal nodes. (b) Computed tomography (CT) scan Although we have found US to be superior to CT scan in identifying metastatic lymph nodes in the central and lateral neck, CT is far superior in the assessment of central neck structures with invasive disease. Additionally, if nodes appear to extend inferiorly by US, we add CT to determine if disease continues into the mediastinum and to what extent. We do not hesitate to use
123
(c)
Operative strategy (a)
Central neck Only about 10 % of our patients required either unilateral or bilateral thyroid resection as part of the reoperative procedure. However, 58 % of our patients required central neck lymph node dissection. We could not ascertain from outside records whether the excised compartment VI nodes were macroscopically evident (i.e., not prophylactically excised) at the initial operation. Depending on either palpable or imaging indications, when technically feasible, we intended to clear the entire side of the central compartment as well as the pretracheal space. This often extended to the innominate artery inferiorly, but metastatic nodes were much less frequently found near the insertion of the RLN than inferiorly. This helped in protecting the area where the superior parathyroid gland would typically be found. Technically, the initial key step for safe dissection is to identify the carotid artery and clear its anterior surface from the thyroid cartilage to the base of the neck. Identification and dissection of the RLN was facilitated by dissecting low in the neck, below the level explored at the previous operation, often indicated by the presence of normal fat. Nerve monitoring was routinely used in about twothirds of the operations, but without any difference in RLN injury when not utilized. Magnification was often helpful. Autotransplantation of devascularized parathyroid glands was used, but most frequently parathyroid glands were not recognized during the dissection, yet the patients maintained normal calcium regulation postoperatively in all but 3 % of patients. We chose not to re-explore a central neck compartment if no indication by palpation or imaging of disease was present in order to avoid the risk of parathyroid devascularization. This is in contrast to the practice advised by Schuff et al. [8] and Steward [9]. (b) Lateral neck Lateral neck dissection was necessitated in 69 % of patients, with about equal proportion of compartments II–V, III–V, and selected single- or
World J Surg (2014) 38:645–652
(c)
two-compartment dissections. Limited dissections were used only if more extensive lateral neck dissections had previously occurred, either by our group, or as apparent when undertaking dissection in the operating room. Prior operative reports were not always accurate in describing the prior extent of dissection. If only compartment IV nodes were identified in a lateral compartment, we did not necessarily extend the superior dissection to encompass compartment II. We did not dissect a central or lateral neck compartment unless there was confirmed evidence of disease. The incisions were determined by surgeon preference, but often did require a vertical component depending on body habitus or ability to safely identify and expose the spinal accessory nerve. In contrast to the carotid sheath structures, phrenic nerve, and thoracic duct that are located on the periphery of the nodal packet, the spinal accessory and cervical plexus nerves run through this packet and need careful dissection to preserve them. Dissection posterior to the carotid artery is rarely necessary and jeopardizes the cervical sympathetic chain. Potentially overlooked are nodes anterior to the internal jugular vein and bifurcation of the carotid artery. Nodes low in the internal jugular chain— compartment IV—are common in PTC, which requires careful dissection to either preserve the thoracic duct, or ensure it is ligated, sometimes with several contributing branches. Operative complications Not surprisingly, the 17 % complication rate in this selected patient series was markedly higher than in first-time thyroid cancer operations. In patients having undergone multiple prior operations for more virulent disease, the more significant complications of hypoparathyroidism and nerve damage were comparable to other reported series [6, 10], but higher than reports of no complications when disease was much more modest [11] or the dissections were limited to the central neck [12].
651
structurally persistent disease (14 %). In fact, nearly 40 % of the reoperative patients suffered relapse subsequent to our reoperation at a mean of 2.1 years, nearly equally split between further nodal recurrence only, or some combination of true soft tissue recurrence and/or systemic disease with or without additional nodal disease. Recognizing that exact characterization between local recurrence and lymph node disease is not perfect, the patients with subsequent lymph node disease relapse we felt deserved careful scrutiny. They have disease that should potentially pose limited threat to life and perhaps be amenable to improved local control. Holding radiologists accountable for any lymph node disease discovered in any compartment at any time subsequent to the reoperation that was not identified on the initial preoperative scan is probably too strict, but there is potential to further reduce relapse in 5–8 % of patients. Even more patients (10 %) relapsed with additional lymph nodes in the previously operated field—more opportunity for improvement. Not analyzed were any subsequent reoperations that were necessary in a proportion of these patients. Even with relapse limited only to lymph node disease, the potential risk of damage to important structures associated with additional reoperation prompted a new avenue of thinking and treatment—US-guided alcohol ablation. The risks, efficacy, and cost analysis of this exciting adjunct intervention have been reported separately [14]. Soft tissue recurrence and/or systemic relapse represents the kind of cancer recurrence more commonly associated with higher-grade malignancies such as breast or colon carcinomas. To identify these higher-risk patients with PTC previously garnered considerable research attention and led to the development of important scoring systems [15]. The prognostic factors making up these scoring systems, such as aggressive histology, large tumor size, age over 45 years, and extra-thyroid extension have also been associated with reduced survival in the reoperative setting [16]. However, death due to PTC was distinctly rare in the usually reported, initially operated series of patients, quite different from this reoperative patient cohort, where such patients are over-represented.
Patient outcomes Conclusions We were gratified that nearly three-quarters of our patients at last follow-up had no structural evidence of disease. However, this is in stark contrast to disease recurrence being prevented in 95 % of patients after initial operation in our experience when optimal principles were followed of preoperative neck US, bilateral thyroidectomy, routine central neck dissection, and lateral neck dissection for disease [13]. Strikingly, 25 % of the reoperative patients had either died of disease (11 %) or were alive with
Although 72 % of reoperative patients can be rendered structurally disease-free after reoperation, nearly 40 % suffer additional relapse. Alterations in surgical technique or improved preoperative localization might affect about 15–20 % of patients; at least 20 % of the patients reflect increased biologic aggressiveness of the disease. The rate of complications, although greater than initial operations, seems acceptable.
123
652
References 1. Cooper D, Doherty G, Haugen B et al (2009) Revised American Thyroid Association management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid 19(11):1167–1214 2. Schlumberger M (1998) Papillary and follicular thyroid carcinoma. N Engl J Med 338(5):297–306 3. Wang T, Dubner S, Sznyter L et al (2004) Incidence of metastatic well-differentiated thyroid cancer in cervical lymph nodes. Arch Otolaryngol Head Neck Surg 130:110–113 4. Robbins K, Shaha A, Medina J et al (2008) Consensus statement on the classification and terminology of neck dissection. Arch Otolaryngol Head Neck Surg 134:536–538 5. Clayman G, Shellenberger T, Ginsberg L et al (2009) Approach and safety of comprehensive central compartment dissection in patients with recurrent papillary thyroid carcinoma. Head Neck 31(9):1152–1163 6. Tufano R, Bishop J, Wu G (2012) Reoperative central compartment dissection for patients with recurrent/persistent papillary thyroid cancer: efficacy, safety, and the association of the BRAF mutation. Laryngoscope 122:1634–1640 7. Farrag TY, Agrawal N, Sheth S et al (2007) Algorithm for safe and effective reoperative thyroid bed surgery for recurrent/persistent papillary thyroid carcinoma. Head Neck 29(12):1069–1074 8. Schuff K, Weber S, Givi B et al (2008) Efficacy of nodal dissection for treatment of persistent/recurrent papillary thyroid cancer. Laryngoscope 118:768–775 9. Steward D (2012) Update in utility of secondary node dissection for papillary thyroid cancer. J Clin Endocrin Metab 97(10):3393–3398
123
World J Surg (2014) 38:645–652 10. Shah M, Harris K, Nassif R et al (2012) Efficacy and safety of central compartment neck dissection for recurrent thyroid carcinoma. Arch Otolaryngol Head Neck Surg 138(1):33–37 11. Al-Saif O, Farrar W, Bloomston M et al (2010) Long-term efficacy of lymph node reoperation for persistent papillary thyroid cancer. J Clin Endocrin Metab 95:2187–2194 12. Alvarado R, Sywak M, Delbridge L et al (2009) Central lymph node dissection as a secondary procedure for papillary thyroid cancer: is there added morbidity? Surgery 145:514–518 13. Grant C, Stulak J, Thompson G et al (2010) Risks and adequacy of an optimized surgical approach to the primary surgical management of papillary thyroid carcinoma treated during 1999–2006. World J Surg 34:1239–1246. doi:10.1007/ s00268-009-0307-9 14. Hay ID, Lee RA, Davidge-Pitts C et al (2013) Long-term outcome of ultrasound-guided percutaneous ethanol ablation of selected ‘‘recurrent’’ neck nodal metastases in 25 patients with TNM stages III or IVA papillary thyroid carcinoma previously treated by surgery and (131)I therapy. Surgery 154(6):1448–1455 15. Hay ID, Bergstralh EJ, Goellner JR et al (1993) Predicting outcome in papillary thyroid carcinoma: development of a reliable prognostic scoring system in a cohort of 1779 patients surgically treated at one institution during 1940 through 1989. Surgery 114:1050–1058 16. Ito Y, Higashiyama T, Takamura Y et al (2011) Prognosis of patients with papillary thyroid carcinoma showing postoperative recurrence to the central neck. World J Surg 35:767–772. doi:10. 1007/s00268-010-0924-3
