Figure. Pattern of lateral neck involvement with metastasis in 248 selective neck dissections.
Merdad M, Eskander A, Kroeker T, Freeman JL. Predictors of Level II and Vb Neck Disease in Metastatic Papillary Thyroid Cancer. Arch Otolaryngol Head Neck Surg. 2012;138(11):1030-1033. doi:10.1001/2013.jamaoto.393
Author Affiliations: Departments of Otolaryngology–Head and Neck Surgery, University of Toronto (Drs Merdad, Eskander, and Freeman), and Mount Sinai Hospital (Drs Kroeker and Freeman), Toronto, Ontario, Canada.
Objective To identify predictors of levels II and Vb involvement in papillary thyroid cancer (PTC) with lateral neck metastasis.
Design Large case series.
Setting High-volume tertiary care hospital.
Patients Consecutive sample of 185 patients who underwent 248 selective neck dissections of at least levels II to V for pathologically proven PTC.
Main Outcome Measures Significant independent predictors of level II and Vb metastasis, including age and pathologic variables (tumor diameter, dominant nodule cellular pathology, multifocality, extracapsular invasion, positive margins, and lymphovascular invasion).
Results Levels II and Vb were involved in 49.3% and 29.2% of our cohort, respectively. Age and lymphovascular invasion were independent predictors of level Vb involvement with metastasis (logistic regression: odds ratio for age = 0.92, SE = 0.03, P = .02; and odds ratio for lymphovascular invasion = 5.52, SE = 0.80, P = .03). No significant predictors were identified for level II involvement.
Conclusions Levels II and Vb were involved in a significant number of patients with PTC and lateral neck disease. Younger age and lymphovascular involvement were independent risk factors for level Vb involvement in patients with PTC and lateral neck metastasis. The increased risk might be of marginal clinical significance. No significant predictors were identified for level II involvement. Our findings do not favor a limited neck dissection on the basis of any of the study's clinical or pathologic variables, and we therefore recommend the routine excision of levels IIa to Vb in all patients with PTC presenting with lateral neck disease.
Papillary thyroid cancer (PTC) commonly spreads to regional lymph nodes.1 Metastasis to the lateral neck largely occurs in a predictable manner, with the middle and lower jugular lymph nodes most commonly involved.2- 6 Several pathologic factors have been previously shown to have a significant association with lateral neck metastasis in PTC, including multifocality and thyroid capsular invasion.2,3
The effect of PTC lymph node metastasis on survival remains debatable, although it is well established that patients with lymph node metastasis are at higher risk for disease recurrence.7 Recent guidelines have recommended an extensive selective neck dissection of at least levels IIa, III, IV, and Vb for patients with PTC and lateral neck disease.8,9 This approach is recommended for all patients with a fine-needle aspiration–proven PTC-positive lymph node detected in the lateral neck. There are no described subgroups in which less surgery may be equally as efficacious in clearing disease.
Lateral neck dissection is associated with a low but not insignificant rate of morbidity. Shoulder dysfunction, from injury to the spinal accessory nerve, and chyle leak are significant complications related to level II and Vb neck dissection. Previously published data,4- 6 including a recent large case series by our group,10 have shown levels II and Vb to be less frequently involved with disease than levels III and IV in metastatic PTC. In this study, we aimed to identify predictors of levels II and Vb involvement in metastatic PTC with the hope of determining factors that may help limit the extent of the required neck resection.
A retrospective medical chart review was conducted for all patients with PTC who underwent lateral neck dissection surgery from January 1, 2004, through January 1, 2011, at Mount Sinai Hospital. The research ethics review board approved the study before commencement.
The study included primary cases (with total or completion thyroidectomy) and recurrent cases presenting with lateral neck disease after an initial thyroid surgery. Our series included patients who underwent unilateral or bilateral neck dissections of at least levels IIa and IIb, III, IV, and Vb (II-Vb). Excluded from this study were cases with less than a selective neck dissection of levels II through Vb, an incomplete pathology report, pathologic features other than PTC, or a history of lateral neck dissection.
Demographic and pathologic data on tumor diameter, dominant nodule cellular features, multifocality, extracapsular invasion, positive margins, and lymphovascular invasion were collected for each patient from the surgical pathology report. In addition, data on neck-level involvement and number of positive lymph nodes per level were collected. Central neck tissue (level VI) was not addressed in this study. All surgical procedures were performed by the senior surgeon (J.L.F.) and his team (fellow and/or resident). All neck specimens were intraoperatively labeled by level for the pathologist. Neck levels and sublevels were defined as previously published,11 with the exception of extending level Vb dissection cranially to the level of the accessory nerve.
In testing univariate associations, t test and Fisher exact test were used based on the type of variable. Logistic regression was used for multivariate analysis. All variables used in univariate analysis were included in the logistic regression model. Statistical significance was defined as P < .05, and the analyses were performed using SPSS Statistics software, version 20 (IBM).
The demographic, clinical, and pathologic data for our patients are presented in Table 1. Levels II, III, IV, and Vb were involved in 49.3%, 76.6%, 61.6%, and 29.2%, respectively (Figure). Classic papillary thyroid carcinoma was the predominant cell pathology (64.9%), followed by the follicular variant subtype (18.1%). Aggressive pathologic variants (eg, tall cell or diffuse sclerosing) were present in 12.9% of our series.
Levels III and IV were concomitantly involved with metastasis in 46.0% of our patients without involvement of other lateral neck levels. Skip metastasis, defined as involvement of levels II or Vb without involvement of levels III and IV, was present in 9.0% of our sample.
Univariate and multivariate analysis of predictors of levels II and Vb involvement are presented in Table 2. None of the studied predictors showed a significant association with metastasis to level II. Age was the only variable showing significant association with level Vb metastasis on univariate analysis (t test, P = .01). On multivariate analysis, younger age and lymphovascular involvement were independent risk factors for level Vb involvement with metastasis (logistic regression, odds ratio for age = 0.92, SE = 0.03, P = .02; and odds ratio for lymphovascular invasion = 5.52, SE = 0.80, P = .03). Similar results were observed when age was dichotomized to younger than 45 years vs 45 years or older. Age younger than 45 years was an independent risk factor for level Vb involvement with metastasis (logistic regression, odds ratio = 0.14, SE = 0.91, P = .03) .
Our study aimed to explore predictors of levels II and Vb disease in metastatic PTC with lateral neck disease. Younger age and lymphovascular involvement were independent risk factors for level Vb involvement in patients with PTC and lateral neck metastasis. No significant predictors were identified for level II involvement. Our results support the current guidelines recommending routine excision of levels IIa to Vb in PTC presenting with lateral neck disease.
Although extensive lateral neck dissection may not have a significant effect on overall survival, it has the potential of decreasing regional recurrences. A previous survival analysis12 by our institution demonstrated a significant decrease in disease-free survival in patients with PTC and lateral neck metastasis. Patients with lateral neck disease had a 6-fold increased chance of disease recurrence compared with patients without lymph node metastasis or with disease limited to the central compartment. The effect of lymph node metastasis on overall survival remains controversial.7,12 Therefore, the approach to PTC with lateral neck disease should focus on removing all levels where disease commonly metastasizes with the least postoperative morbidity possible.
The most recent guidelines concerning the management of lateral neck disease in metastatic PTC favor extensive neck dissections. The practice of “berry picking” has been shown to have a much higher recurrence rate compared with a comprehensive neck dissection targeting levels commonly involved in metastatic PTC.13 The American Thyroid Association's consensus review8 on the rationale for lateral neck dissection in differentiated thyroid cancer recommends the routine dissection of levels IIa, III, IV, and Vb when metastatic disease is detected at any of the lateral neck levels. Similarly, the Triological Society Best Practice guidelines9 recommend a selective neck dissection targeting levels IIa through Vb with or without the inclusion of levels IIb and Va (on the basis of the presence of suspicious disease at these levels).
A “one size fits all” approach, as recommended by recent guidelines, has the potential of overtreating a number of patients in which metastasis might be limited to 2 or 3 lymph node levels. Almost half the cohort in our present study had disease limited to levels III and IV. Dissection of level II, and the manipulation of the spinal accessory nerve, is associated with significant postoperative shoulder complaints.14- 18 Transection of cervical nerve rootlets in level V surgery has been associated with postoperative numbness, neuropathic pain, and decreased quality of life.19,20 Furthermore, the risk of chyle leak is not insignificant in a comprehensive level IV and V neck dissection.21
A highlyselective neck dissection (limited to levels III and IV) would be advocated in patients with PTC and lateral neck disease if (1) the rate of metastasis to levels II and Vb was shown to be low, (2) it was possible to predict, on the basis of certain variables, which patients are at higher risk of involvement at these levels, and (3) the increased risk associated with these variables was clinically significant. In our cohort, levels II and Vb were involved with disease in a significant number of cases: 49.3% in level II and 29.2% in level Vb. These relatively high rates of involvement with metastasis are similar to those found by others.2- 6 On the basis of our logistic regression model, younger age and lymphovascular involvement were independent predictors of level Vb metastatic involvement. The slight increased risk in younger patients is probably of minimal clinical significance. None of the studied variables had a significant association with level II metastasis. Our findings do not favor a limited neck dissection on the basis of any of the study's clinical or pathologic variables. We therefore recommend a routine comprehensive selective neck dissection of levels II, III, IV, and Vb for all patients with PTC and any lateral neck disease.
Farrag et al4 previously published an article on predictors of level IIb lymph node metastasis in patients with PTC. Age, sex, N category, and other levels of cervical lymph node involvement were studied, and none of the variables were found to significantly predict level IIb involvement with disease. Koo et al3 observed 76 patients with PTC and lateral neck disease and found the concurrent involvement of levels IIa, III, and IV to be an independent predictor of level IIb involvement, whereas there was no significant association found with age, sex, tumor size, multifocality, lymphovascular invasion, and capsular invasion. Farrag et al,4 Caron et al,22 and Koo et al3 have demonstrated level IIb disease to occur with a much higher frequency when level IIa is involved. In our series, levels IIa and IIb were routinely excised in all patients and were labeled to the pathologist as a single group (level II). Therefore, our analysis was limited to predicting level II metastasis without sublevel analysis. In our opinion, the distinction between levels IIa and IIb is to some extent arbitrary, especially when there is disease involvement at these levels. By routinely dissecting level IIb, we hope to decrease the rate of recurrence and the need for any future dissection around the spinal accessory nerve in a previously operated-on neck. The risk of recurrence at level IIb has been shown to be as high as 20%.22
On the basis of data from a cohort of 70 patients, Kupferman et al2 examined predictors of level V involvement in PTC with lateral neck disease. Level V involvement was associated with multifocality and level II, III, and IV metastasis, whereas there was no significant association with age, sex, tumor size, and pathologic subtype. Involvement of level V in PTC has been shown to be restricted to lymph nodes below the spinal accessory nerve. In our cohort, dissection of level V included sublevel Vb and the part of Va below the spinal accessory nerve.23
There is an evident inconsistency in the literature regarding predictors of lateral neck–level involvement in metastatic PTC. This could be attributed to the relatively small sample size of studies addressing the issue, including our present study. A large sample size is required to accurately detect significant predictors, and this in turn requires a multi-institutional or international collaboration. Furthermore, surgical techniques and pathologic review standards are currently fairly heterogeneous, which could significantly compromise the results of such collaboration. Another limitation of our study is the lack of follow-up data with regard to recurrence and survival. In our series, 43.8% of the patients initially presented with recurrence of PTC in the lateral neck. These patients may have received radioactive iodine treatment after their original thyroidectomy, and the effect of radioactive iodine treatment on the sensitivity of pathologic detection of disease in the neck dissection specimen is unknown.
In summary, levels II and Vb were involved in a significant number of patients with PTC and lateral neck disease. Younger age and lymphovascular invasion are independent risk factors for level Vb involvement in patients with PTC and lateral neck metastasis. No significant predictors were identified for level II involvement. We recommend the routine excision of levels IIa to Vb in all patients with PTC presenting with any lateral neck disease.
Correspondence: Mazin Merdad, MD, MPH, Department of Otolaryngology–Head and Neck Surgery, University of Toronto, 190 Elizabeth St, Room 3S438, Toronto, Ontario M5G 2N9, Canada (firstname.lastname@example.org).
Submitted for Publication: July 6, 2012; final revision received July 24, 2012; accepted August 16, 2012.
Author Contributions: Dr Merdad had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Merdad and Freeman. Acquisition of data: Merdad. Analysis and interpretation of data: Merdad, Eskander, and Kroeker. Drafting of the manuscript: Merdad and Eskander. Critical revision of the manuscript for important intellectual content: Eskander, Kroeker, and Freeman. Statistical analysis: Merdad. Administrative, technical, and material support: Eskander. Study supervision: Kroeker and Freeman.
Conflict of Interest Disclosures: None reported.
Additional Contributions: We thank Spiro J. Eski, MD, for help with data collection.