Computed Tomography for Detecting Cervical Lymph Node Metastasis in Patients Who Have Papillary Thyroid Microcarcinoma with Tumor Characteristics Appropriate for Active Surveillance

Abstract

Background:

Active surveillance (AS) has been considered one of the management options in patients with low-risk papillary thyroid microcarcinoma (PTMC). It is important to evaluate clinical lymph node (LN) metastasis to select appropriate candidates with low-risk disease. We investigated the predictive accuracy of computed tomography (CT) for cervical LN metastasis in patients who have PTMC with tumor characteristics appropriate for AS.

Methods:

This was a retrospective study. Medical records from December 2014 to the end of 2016 were reviewed. Patients who underwent thyroidectomy and who had pathologically confirmed PTMC were included. A total of 464 patients who had tumors with ultrasound (US) characteristics appropriate for AS and who underwent preoperative CT were included in the analysis.

Results:

CT showed higher diagnostic values especially in positive predictive value (PPV) than US. In patient-based analyses, CT showed low sensitivity and negative predictive value (NPV) (16.0% and 58.5%, respectively), but high specificity and PPV (99.6% and 97.1%, respectively) for detecting cervical LN metastasis. Similar trends were observed for the results of the central neck-level by CT (sensitivity, 14.9%; specificity, 97.4%; PPV, 82.9%; and NPV, 57.4%) in level-by-level analyses. When restricted to lateral neck levels, CT showed high diagnostic accuracy of 95.4% for detecting LN metastasis. In all analyses, CT showed better diagnostic values for cervical LN metastasis than US. Combining US and CT did not improve the diagnostic accuracy compared with CT.

Conclusions:

In patients with PTMC whose tumor has characteristics suitable for AS, CT had additional benefit after cervical LN assessment by US. Further studies are needed to evaluate routine initial CT scanning for patients who are candidates for AS.

Introduction

Papillary thyroid carcinoma (PTC) is the most common type of thyroid malignancy, accounting for more than 80% of cases (1). Among PTCs, tumors measuring ≤10 mm are defined as papillary thyroid microcarcinoma (PTMC). Diagnoses of PTMC have increased with the improvements in imaging tools, resulting in a changing incidence of PTC (2 –4). Many previous studies have demonstrated that in most patients with low-risk PTMC, the disease does not progress during active surveillance (AS) (5 –8). Based on these results, AS was adopted as a treatment option in the Japanese guidelines of 2011 and by the American Thyroid Association in 2015 (9,10). In these guidelines, low-risk tumors are defined as PTMC without clinically evident metastases or local invasion, and with no convincing cytologic evidence of aggressive disease (10).

A recent study by Brito et al. (11) suggested a clinical framework for selecting patients with PTMC who are suitable for AS. This framework defined the characteristics of tumor and neck ultrasound (US) findings that were appropriate for AS. In particular, the authors suggested that clinical N1 disease was inappropriate for AS. In other words, it is important to assess the risk category of the patient's disease before starting AS, and especially to assess whether there is metastasis to cervical lymph nodes (LNs), the most common site of metastasis in PTC patients.

According to the American Thyroid Association guidelines, preoperative neck US is recommended to evaluate the cervical LNs. Other neck imaging modalities such as computed tomography (CT), magnetic resonance imaging, and positron emission tomography are recommended as an adjunct to US (10). Although some studies have compared the diagnostic accuracy for LN metastasis of different imaging modalities (12 –15), the results are inconsistent. To our knowledge, there has been no study comparing the diagnostic accuracy of CT and US for LN metastasis in patients with PTMC in terms of screening candidates for AS. In this study, we investigated the accuracy of CT for detecting cervical LN metastasis in patients who had PTMC with tumor characteristics appropriate for AS.

Subjects and Methods

Study subjects

This was a retrospective study and the study protocol was approved by the Institutional Review Board of Seoul National University Bundang Hospital (SNUBH) (approval no.: 101-408-107). We reviewed the medical records of patients who underwent thyroidectomy and who had pathologically confirmed PTMC from December 2014 to the end of 2016. All patients were treated by one surgeon (J.Y.C.) to reduce potential bias.

The patients who underwent preoperative CT and who had tumors with US characteristics appropriate for AS, as suggested by Brito et al. (11) and applied to an ongoing prospective AS study (16), were selected. Tumor characteristics appropriate for AS included no evidence of aggressive cytology on the fine-needle aspirate, not located in the subcapsular area adjacent to the recurrent laryngeal nerve (RLN), no evidence of extrathyroidal extension, no clinical evidence of invasion of RLN or trachea, and no clinical M1 disease.

Finally, 464 patients were included in the analyses. Because of a lack of information about LN characteristics for US that was performed outside SNUBH, only 218 patients who underwent both US and CT in SNUBH were included in the analysis to compare the diagnostic performance of US and CT for detecting cervical LN metastasis.

US examinations

All US examinations were performed using a 12 MHz linear probe US system (iU22 or Epiq 5; Philips Medical Systems, Bothell, WA). Three experienced radiologists conducted the US examinations for assessment of thyroid nodules and central and lateral neck lymphadenopathy. Thyroid nodules were evaluated for shape, composition, echogenicity, margin, orientation, vascularity, and presence of calcification (17). Disruption of the thyroid capsule by the PTMC nodule or any evidence of tumor invasion of adjacent organs (e.g., esophagus, nerves, trachea, major vessels, or muscle) was defined as extrathyroidal extension.

LNs were categorized in three grades: probably benign, indeterminate, or suspicious for metastasis as defined by the consensus statement of the Korean Society of Thyroid Radiology (17). Benign LNs were defined as LNs showing either a central echogenic hilum or central hilar vascularity. Suspicious LNs were LNs with any one suspicious finding, including enlargement >1 cm in the short diameter, cystic change, hyperechogenicity, peripheral or diffuse vascularity, and calcification. LNs showing neither suspicious nor benign US features were classified as indeterminate.

Neck CT

Multidetector CT systems (Brilliance 64 or iCT 256; Philips Medical Systems, Bothell, WA) were used for all CT scans. All CT scans included precontrast and postcontrast scans from the skull base to carina using 120 kVp and 250 mAs. Multiplanar reconstruction was performed for axial, coronal, and sagittal planes with 2-mm thickness and 2-mm increments. Suspicious LNs (positive finding on CT) were defined as LNs with any one suspicious CT finding, including enlargement >1 cm in the short diameter, cystic change, strong enhancement, heterogeneous enhancement, and calcification.

Surgery

All patients with PTMC were treated with total thyroidectomy (n = 172) or lobectomy (n = 292) with prophylactic bilateral or ipsilateral central neck dissection (CND) according to their disease status (Table 1). Total thyroidectomy was performed when bilateral PTMC or LN metastasis was confirmed before surgery. In cases of confirmed metastasis to the lateral LN in the preoperative biopsy or the frozen section obtained during surgery, a modified radical neck dissection (MRND) on the ipsilateral side was performed. During thyroid lobectomy and total thyroidectomy, the surgeon attempted to preserve the RLN and upper and lower parathyroid glands to reduce postoperative complications.

Table 1.

Baseline Characteristics of Study Subjects (n = 464)

Demographic and biochemical parameters
Age, years	45.9 ± 10.8
Female	357 (76.9)
Height, cm	162.2 ± 8.4
Weight, kg	64.0 ± 12.5
BMI, kg/m²	24.2 ± 3.6
TSH, μIU/L	1.76 (1.12–2.55)
fT4, ng/dL	1.12 (1.04–1.26)
Total T3, ng/dL	124 (114–134)

Tumor characteristics
Maximum tumor diameter, mm	7.1 ± 1.9
Number of tumors
1	335 (72.2)
2	71 (15.3)
3	33 (7.1)
4	25 (5.4)
Location of tumor
Right	200 (43.1)
Left	200 (43.1)
Both	64 (13.8)

LN characteristics
No LN metastasis	236 (50.9)
Central LN metastasis	206 (44.4)
Lateral LN metastasis	22 (4.7)

Extent of surgery
Total thyroidectomy	172 (37.1)
CND, ipsilateral	68 (14.7)
CND, ipsilateral+MRND, ipsilateral	4 (0.9)
CND, bilateral	77 (16.6)
CND, bilateral+MRND, ipsilateral	22 (4.7)
CND, bilateral+MRND, bilateral	1 (0.2)
Lobectomy	292 (62.9)
CND, ipsilateral	267 (57.5)
CND, ipsilateral+MRND, ipsilateral	22 (4.7)
CND, bilateral	3 (0.6)

Values are expressed as mean ± standard deviation, n (%), or median (interquartile range).

BMI, body mass index; CND, central neck dissection; LN, lymph node; MRND, modified radical neck dissection; T3, triiodothyronine; fT4, free thyroxine; TSH, thyrotropin.

Pathology analysis

Surgical pathology reports included the pathologic characteristics of the tumor as follows: (i) type of surgery, (ii) location and size of tumor, (iii) histopathologic subtype of PTC, (iv) lymphatic and blood vessel invasion, (v) extrathyroidal extension, (vi) surgical margins, (vii) LNs, (viii) the largest dimension of metastatic carcinoma in LNs, and (ix) pathologic TNM staging. The neck was divided into levels bilaterally from levels II to VI according to an imaging based nodal classification (18).

From the pathology reports, we reviewed the location and size of the tumor and the LN metastasis. For the analyses, size of LN metastasis was categorized as micrometastasis (<2.0 mm), small nodal metastasis (2.0–9.9 mm), and macrometastasis (≥10 mm) (19,20). It was difficult to match “per node” between pathology and imaging results, so analysis was performed “per level” according to the nodal classification. The size information of LN metastasis was missing in nine patients.

Image analysis

An experienced radiologist (Y.K.K.) reviewed all US and CT images stored in a picture archiving and communication system of SNUBH retrospectively. US images were reviewed to determine the PTMC nodules that were eligible for AS. The eligibility criteria were applied from a suggested clinical framework for AS and an ongoing prospective AS study as mentioned above (11,16). US and CT images were reviewed for evaluation of central and lateral LNs. Positive findings by each imaging study were defined as showing suspicious LNs. To reduce bias, the radiologist was blinded to the pathologic results. The list of patients was randomized for CT review, and the CT image review was performed 5 months after the US review to avoid recall bias.

Statistical analyses

Continuous data are reported as mean ± standard deviation or median (interquartile range), as appropriate. Categorical data are reported as n (%). The final diagnosis of LNs was determined by pathology readings of surgical specimens. The sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV), and diagnostic accuracy of the imaging modalities (US and CT) for detecting cervical LN metastases were compared with the pathology results. Patient-based and level-by-level analyses of diagnostic values were performed. Since the size of LN metastasis was reported as its largest dimension, subgroup analysis according to the size of LN metastasis was able to be conducted in patient-based analysis only. All statistical analyses were performed using SPSS for Windows v 22.0 (IBM Corp., Armonk, NY). Significance was set at p < 0.05.

Results

Baseline characteristics

A total of 464 patients (357 women; mean age, 45.9 ± 10.8 years) with PTMC were included in the analysis. Patients and tumor characteristics are shown in Table 1. The average tumor size was 7.1 ± 1.9 mm and 72.2% of the patients had a single tumor. Bilateral tumors were found in 64 patients (13.8%).

Diagnostic value of CT for cervical LN metastasis

Table 2 demonstrates the diagnostic results of CT for predicting the presence of cervical LN metastasis. Most of the metastatic LNs not detected by CT were micrometastases or small metastases on pathology. The median size of the LN metastases was 2.0 mm (1.0–3.0) [range 0.1–15.0 mm] and the median number of the metastatic LNs per patient was 2 (1 –3) [range 1–26].

Table 2.

Concordance Between Computed Tomography and Pathologic Results for Cervical Lymph Node Metastasis

	CT
	Test (−)	Test (+)
Patients (n = 464^*)
Pathology (−)	251	1
Pathology (+)	178	34
Micrometastasis (<2.0 mm)	76	4
Small nodal metastasis (2.0–9.9 mm)	90	19
Macrometastasis (≥10.0 mm)	3	11
All neck levels (n = 1000)
Pathology (−)	718	14
Pathology (+)	210	58
Central neck levels (n = 496)
Pathology (−)	261	7
Pathology (+)	194	34
Lateral neck levels (n = 504)
Pathology (−)	457	7
Pathology (+)	16	24

Nine patients were excluded subgroup analysis due to lack of the size information of LN metastasis.

CT, computed tomography.

The sensitivity, specificity, PPV, NPV, and accuracy are presented in Table 3. Although CT showed an accuracy of 61.4%, it showed a high specificity and PPV (99.6% and 97.1%, respectively) in patient-based analyses. In a subgroup analysis considering the size of LN metastasis, the sensitivity of CT was improved with incremental size (5% in micrometastasis, 17.4% in small nodal metastasis, and 78.6% in macrometastasis).

Table 3.

Diagnostic Values of Computed Tomography for Cervical Lymph Node Metastasis

	Diagnostic values
	Sensitivity (%)	Specificity (%)	PPV (%)	NPV (%)	Accuracy (%)	LR+	LR−
Patients (n = 464^*)	16.0	99.6	97.1	58.5	61.4	40.0	0.84
Micrometastasis (<2.0 mm)	5.0
Small nodal metastasis (2.0–9.9 mm)	17.4
Macrometastasis (≥10.0 mm)	78.6
All neck levels (n = 1000)	21.6	98.1	80.6	77.4	77.6	11.4	0.80
Central neck levels (n = 496)	14.9	97.4	82.9	57.4	59.5	5.7	0.87
Lateral neck levels (n = 504)	60.0	98.5	77.4	96.6	95.4	40.0	0.41

Nine patients were excluded subgroup analysis due to lack of the size information of LN metastasis.

LR−, negative likelihood ratio; LR+, positive likelihood ratio; NPV, negative predictive value; PPV, positive predictive value.

Level-by-level analyses showed different results for the central and lateral neck levels. Although the specificity was high for both neck levels (central, 97.4%; lateral 98.5%), the sensitivity was lower for central neck levels (14.9%) and higher for lateral neck levels (60.0%). The PPV was higher than the NPV for central neck levels (PPV, 82.9%; NPV, 57.4%), whereas for lateral neck levels, the NPV was higher than the PPV (PPV, 77.4%; NPV, 96.6%). The diagnostic accuracy of CT for detecting LN metastasis in lateral neck levels was 95.4%, whereas it was 59.5% for central neck levels.

Comparison of diagnostic value of US, CT, and combined US/CT

Among the study subjects, 218 patients who underwent both preoperative US and CT at SNUBH were included in the analysis. Because the sensitivity of both US and CT was low, to improve sensitivity, we defined a positive result for combined US and CT (US+CT) when either modality gave a positive result. The diagnostic results of each examination are presented in Table 4.

Table 4.

Concordance Between Imaging Tests (Ultrasound, Computed Tomography, Ultrasound+Computed Tomography) and Pathologic Results for Cervical Lymph Node Metastasis in Patients Who Underwent Both Ultrasound and Computed Tomography (n = 218)

	US		CT		US+CT
	Test (−)	Test (+)	Test (−)	Test (+)	Test (−)	Test (+)
Patients (n = 218^*)
Total
Pathology (−)	106	12	118	0	106	12
Pathology (+)	88	12	81	19	77	23
Micrometastasis (<2.0 mm)	35	3	34	4	32	6
Small nodal metastasis (2.0–9.9 mm)	47	6	45	8	43	10
Macrometastasis (≥10.0 mm)	4	3	0	7	0	7
All neck levels (n = 479)
Pathology (−)	314	37	346	5	310	41
Pathology (+)	112	16	92	36	87	41
Central neck levels (n = 234)
Pathology (−)	118	11	127	2	117	12
Pathology (+)	98	7	85	20	82	23
Lateral neck levels (n = 245)
Pathology (−)	196	26	219	3	193	29
Pathology (+)	14	9	7	16	5	18

Positive results on US or CT were considered positive.

Nine patients were excluded subgroup analysis due to lack of the size information of LN metastasis.

US, ultrasound; US+CT, combined results of US and CT.

Table 5 shows the diagnostic accuracy of US, CT, and US+CT for detecting cervical LN metastasis. In patient-based analysis, subgroup analysis according to the size of LN metastasis showed similar results with those of all participants. In both patient-based analyses and level-by-level analyses, all diagnostic values were higher for CT than for US. Although US+CT showed a slightly higher sensitivity than CT, the specificity and PPV of US+CT were much lower than those of CT in all analyses. The diagnostic accuracy of US, CT, and US+CT for predicting cervical LN metastasis was 54.1%, 62.8%, and 59.2%, respectively, in patient-based analyses. In level-by-level analyses, the diagnostic accuracy of each modality was higher for lateral neck levels than for central neck levels (lateral neck levels: US, 83.7%; CT, 95.9%; US+CT, 86.1%).

Table 5.

Diagnostic Values for Cervical Lymph Node Metastasis According to Imaging Tests

	Diagnostic values
	Sensitivity (%)	Specificity (%)	PPV (%)	NPV (%)	Accuracy (%)	LR+	LR−
Patients (n = 218^*)
US	12.0	89.8	50.0	54.6	54.1	1.72	0.98
Micrometastasis (<2.0 mm)	7.8
Small nodal metastasis (2.0–9.9 mm)	11.3
Macrometastasis (≥10.0 mm)	42.9
CT	19.0	100.0	100.0	59.3	62.8	—	0.81
Micrometastasis (<2.0 mm)	10.5
Small nodal metastasis (2.0–9.9 mm)	15.1
Macrometastasis (≥10.0 mm)	100.0
US+CT	23.0	89.8	65.7	57.9	59.2	2.3	0.86
Micrometastasis (<2.0 mm)	15.8
Small nodal metastasis (2.0–9.9 mm)	23.3
Macrometastasis (≥10.0 mm)	100.0
All neck levels (n = 479)
US	12.5	89.5	30.2	73.7	68.9	1.2	0.98
CT	28.1	98.6	87.8	79.0	79.7	20.1	0.73
US+CT	32.0	88.3	50.0	78.1	73.3	2.7	0.77
Central neck levels (n = 234)
US	6.7	91.5	25.0	54.6	53.4	0.8	1.02
CT	19.0	98.4	90.9	59.9	62.8	11.9	0.82
US+CT	21.9	90.7	65.7	58.8	59.8	2.4	0.86
Lateral neck levels (n = 245)
US	36.0	88.3	25.7	93.3	83.7	2.1	0.73
CT	69.6	98.6	84.2	96.9	95.9	49.7	0.31
US+CT	78.2	86.9	38.3	97.4	86.1	6.0	0.25

Positive results on US or CT were considered positive.

Nine patients were excluded subgroup analysis due to lack of the size information of LN metastasis.

Diagnostic value of CT as adjunctive modality to US

Table 6 shows the diagnostic results of CT as an adjuvant to US in PTMC patients whose tumor characteristics were appropriate for AS. In 24 patients who had suspicious LNs on US, CT demonstrated a PPV of 100% and an NPV of 83.3%. Of the 194 patients without suspicious LNs on US, all 11 patients who had suspicious LNs on CT proved to have cervical LN metastasis (100% PPV). In these patients, additional CT examination had an NPV of 60.3%.

Table 6.

Concordance Between Computed Tomography and Pathologic Results and Diagnostic Values of Computed Tomography for Cervical Lymph Node Metastasis According to the Results of Ultrasound

	CT in positive US (n = 24)		CT in negative US (n = 194)
	Test (−)	Test (+)	Test (−)	Test (+)
Pathology (−)	12	0	106	0
Pathology (+)	4	8	77	11
Sensitivity (%)	66.7		12.5
Specificity (%)	100.0		100.0
PPV (%)	100.0		100.0
NPV (%)	83.3		60.3

CT in positive/negative US, subgroup analysis of accuracy of CT in patients with positive/negative results in US.

Discussion

In this study, preoperative evaluation of patients with PTMC tumor characteristics appropriate for AS showed that CT had low sensitivity and NPV, but high specificity and PPV for detecting cervical LN metastasis. Prophylactic CND was routinely performed in all patients in our study, and this policy probably caused the low sensitivity and NPV of preoperative imaging because of the increased detection of microscopic LN metastasis.

Previous studies evaluating the diagnostic accuracy of CT for cervical LN metastasis in patients with PTC reported relatively higher sensitivity (50%) and NPV (74%) in central neck levels than we found in our study (14,15). In those studies, CND was not performed routinely but was based on the results of preoperative imaging (14,15). For the lateral neck levels, we performed MRND only in cases where the preoperative biopsy or frozen section obtained during surgery confirmed metastasis to the lateral LN, and our results showed a higher sensitivity (60.0%) and NPV (96.6%) for CT, a finding that is consistent with those of previous studies (14,15).

Although both our study and previous studies (12,14,15,21) revealed better diagnostic accuracy for the lateral neck levels than for the central neck levels, the fact that MRND or lateral neck dissection is performed much more selectively than CND should be considered when interpreting these results. In the present study, patient-based analyses showed similar results to those for central neck levels in level-by-level analyses. This might be because only 11% of our patients underwent MRND.

The high PPV of CT for predicting cervical LN metastasis in our study is noteworthy. A number of studies have reported that disease prognosis in PTC is associated with radiologically identified clinical LN metastasis rather than with microscopic metastasis (20,22 –24). In terms of the pathologic spectrum of LN metastasis based on size, micrometastasis (<2.0 mm) and small nodal metastasis (2.0–9.9 mm) are categorized lower risk N1 disease (< 5% risk of recurrence) (19). Therefore, it is most important to detect clinical LN metastasis in patients with PTMC to select appropriate candidates for AS.

The low NPV of imaging for detecting nodal disease is acceptable in PTC because microscopic LN metastasis has a relatively low prognostic significance and most of the false-negative cases are microscopic LN metastasis. Our results also reveal that 98% of false negative results of CT concerned micrometastases or small nodal metastases in patient-based analysis. However, because nodal disease detectable by imaging reflects prognostically significant clinical LN metastasis, the PPV should be high enough to prevent unnecessary surgical treatment, especially in PTMC patients who may be candidates for AS.

In the analysis of the 218 patients who underwent LN evaluation with both US and CT, all diagnostic values for predicting the presence of cervical LN metastasis were higher for CT than for US. Moreover, combining US and CT did not improve the diagnostic accuracy of CT. In previous studies, US and CT were reported to have similar diagnostic values for predicting cervical LN metastasis (12,14,15).

The inconsistency between our results and those of previous studies may be explained by different study populations. Our study included only patients with PTMC whose tumor characteristics were appropriate for AS, whereas previous studies enrolled all PTC patients regardless of their tumor size or characteristics (12,14,15). The patients in our cohort might be at an earlier disease stage compared to the patients in previous studies, and macroscopic LN metastasis, which can be detected by preoperative imaging, was less frequent than that in previous studies. This difference could amplify the superiority of CT for detecting LN metastasis in the present study.

We also evaluated the diagnostic value of CT as an adjuvant imaging modality after US. When suspicious LNs were found on US, CT had benefits as a confirmatory test (PPV of 100.0%) and additional value, although insufficient value as a rule-out test (NPV of 83.3%). For the 194 patients who had no suspicious LN on US, CT identified 11 individuals who had suspicious LN, and all of whom proved to have nodal disease (PPV of 100.0%). These results might be achievable because CT is able to assess the whole neck compartment regardless of adjacent structures, whereas US has a limited scanning range to evaluate LNs in the retropharyngeal space, mediastinum, and lower level VI (25). Therefore, CT can be an excellent modality in itself and can complement the diagnostic power of US for making a decision about AS in PTMC patients.

This study has some significant clinical implications. It is the first study to evaluate the diagnostic accuracy of CT for detecting cervical LN metastasis in patients with PTMC whose tumor characteristics are appropriate for AS. To date, neck US has been regarded as the essential diagnostic modality to evaluate the characteristics of the tumor in patients with thyroid cancer (10), and examination of the cervical LN to detect metastases is also conducted during neck US. The results of this study suggest that CT has a higher diagnostic accuracy than US for detecting metastatic LNs. However, careful consideration is needed before performing routine CT for making a decision about AS. Although CT showed a high PPV for cervical LN metastasis, it is unclear how helpful CT will be in actual clinical practice.

In long-term observational trials conducted in Japan, patients with PTMC, who had no evidence of cervical LN metastasis on initial US, were included in AS without an initial CT scan (7,8). In these studies, the prognosis of the patients who underwent surgery during AS because their tumor had progressed was as good as that of the patients who underwent surgery immediately after diagnosis (8). Therefore, it is unclear whether potential LN metastasis that is not detected on US but detected on CT will affect the long-term prognosis under an appropriate AS protocol for detecting disease progression.

However, according to our results, there is a possibility that initial CT before the treatment decision for patients with low-risk PTMC may decrease the rate of disease progression during AS apart independent on the long-term prognosis. If so, we could consider modifying the current follow-up protocol for patients who have no suspicious LNs on both US and CT even if AS can be applied to the patients who have no suspicious LNs on US regardless of their CT results. This is one of the reasons to expect the results of an ongoing multicenter prospective study of AS in patients with low-risk PTMC that includes US and CT for initial evaluation (16).

There are some limitations to our study. First, the study design is retrospective and there is a possibility of selection bias. Second, LN analysis was performed on a neck-level basis, and a node-by-node analysis was not performed. Therefore, there is a possibility that the suspicious LNs on US or CT did not exactly match the actual metastatic LN for which pathology results were available.

In conclusion, in patients with PTMC whose tumor characteristics are suitable for AS, CT showed higher diagnostic values than US for detecting metastatic LNs and had additional benefit after cervical LN assessment by US. However, further studies are warranted to evaluate the cost-effectiveness of routine CT scanning for candidates with PTMC for AS in a clinical setting.

Footnotes

Author Disclosure Statement

None of the authors declare potential conflicts of interest relevant to this article.

Funding Information

No funding was received for this article.

References

Sherman

. 2003. Thyroid carcinoma. Lancet (London, England), 361:501–511.

Brito

, Morris

, Montori

. 2013. Thyroid cancer: zealous imaging has increased detection and treatment of low risk tumours. BMJ, 347:f4706.

Castro

, Morris

, Ryder

, Brito

, Hay

. 2015. Most patients with a small papillary thyroid carcinoma enjoy an excellent prognosis and may be managed with minimally invasive therapy or active surveillance. Cancer, 121:3364–3365.

Vaccarella

, Franceschi

, Bray

, Wild

, Plummer

, Dal Maso

. 2016. Worldwide thyroid-cancer epidemic? The increasing impact of overdiagnosis. N Engl J Med, 375:614–617.

Ito

, Uruno

, Nakano

, Takamura

, Miya

, Kobayashi

, Yokozawa

, Matsuzuka

, Kuma

, Miyauchi

. 2003. An observation trial without surgical treatment in patients with papillary microcarcinoma of the thyroid. Thyroid, 13:381–387.

Sugitani

, Toda

, Yamada

, Yamamoto

, Ikenaga

, Fujimoto

. 2010. Three distinctly different kinds of papillary thyroid microcarcinoma should be recognized: our treatment strategies and outcomes. World J Surg, 34:1222–1231.

Ito

, Miyauchi

, Inoue

, Fukushima

, Kihara

, Higashiyama

, Tomoda

, Takamura

, Kobayashi

, Miya

. 2010. An observational trial for papillary thyroid microcarcinoma in Japanese patients. World J Surg, 34:28–35.

Ito

, Miyauchi

, Kihara

, Higashiyama

, Kobayashi

, Miya

. 2014. Patient age is significantly related to the progression of papillary microcarcinoma of the thyroid under observation. Thyroid, 24:27–34.

Takami

, Ito

, Okamoto

, Yoshida

. 2011. Therapeutic strategy for differentiated thyroid carcinoma in Japan based on a newly established guideline managed by Japanese Society of Thyroid Surgeons and Japanese Association of Endocrine Surgeons. World J Surg, 35:111–121.

10.

Haugen

, Alexander

, Bible

, Doherty

, Mandel

, Nikiforov

, Pacini

, Randolph

, Sawka

, Schlumberger

, Schuff

, Sherman

, Sosa

, Steward

, Tuttle

, Wartofsky

. 2016. 2015 American Thyroid Association Management Guidelines for adult patients with thyroid nodules and differentiated thyroid cancer: the American Thyroid Association Guidelines Task Force on Thyroid Nodules and Differentiated Thyroid Cancer. Thyroid, 26:1–133.

11.

Brito

, Ito

, Miyauchi

, Tuttle

. 2016. A clinical framework to facilitate risk stratification when considering an active surveillance alternative to immediate biopsy and surgery in papillary microcarcinoma. Thyroid, 26:144–149.

12.

Ahn

, Lee

, Yi

, Shong

, Hong

, Lee

, Choi

, Kim

. 2008. Diagnostic accuracy of CT and ultrasonography for evaluating metastatic cervical lymph nodes in patients with thyroid cancer. World J Surg, 32:1552–1558.

13.

Choi

, Kim

, Kwak

, Kim

, Chang

, Kim

. 2009. Preoperative staging of papillary thyroid carcinoma: comparison of ultrasound imaging and CT. AJR Am J Roentgenol, 193:871–878.

14.

Lesnik

, Cunnane

, Zurakowski

, Acar

, Ecevit

, Mace

, Kamani

, Randolph

. 2014. Papillary thyroid carcinoma nodal surgery directed by a preoperative radiographic map utilizing CT scan and ultrasound in all primary and reoperative patients. Head Neck, 36:191–202.

15.

Kim

, Park

, Son

, Kim

, Jeon

, Na

. 2008. Preoperative diagnosis of cervical metastatic lymph nodes in papillary thyroid carcinoma: comparison of ultrasound, computed tomography, and combined ultrasound with computed tomography. Thyroid, 18:411–418.

16.

Moon

, Kim

, Lee

, Kong

, Kim

, Jung

, Lee

, Yoo

, Hwangbo

, Song

, Kim

, Cho

, Kim

, Jung

, Choi

, Ryu

, Lee

, Hah

, Jung

, Ryu

, Hwang

, Park

, Sung

, Yi

, Park

. 2018. Study protocol of multicenter prospective cohort study of active surveillance on papillary thyroid microcarcinoma (MAeSTro). Endocrinol Metab (Seoul), 33:278–286.

17.

Shin

, Baek

, Chung

, Ha

, Kim

, Lee

, Lim

, Moon

, Na

, Park

, Choi

, Hahn

, Jeon

, Jung

, Kim

, Kwak

, Lee

, Park

, Sung

, Korean Society of Thyroid Radiology, Korean Society of Radiology. 2016. Ultrasonography diagnosis and imaging-based management of thyroid nodules: revised Korean Society of Thyroid Radiology Consensus Statement and Recommendations. Korean J Radiol, 17:370–395.

18.

Som

, Curtin

, Mancuso

. 2000. Imaging-based nodal classification for evaluation of neck metastatic adenopathy. AJR Am J Roentgenol, 174:837–844.

19.

Randolph

, Duh

, Heller

, LiVolsi

, Mandel

, Steward

, Tufano

, Tuttle

, American Thyroid Association Surgical Affairs Committee's Taskforce on Thyroid Cancer Nodal Surgery. 2012. The prognostic significance of nodal metastases from papillary thyroid carcinoma can be stratified based on the size and number of metastatic lymph nodes, as well as the presence of extranodal extension. Thyroid, 22:1144–1152.

20.

Cranshaw

, Carnaille

. 2008. Micrometastases in thyroid cancer. An important finding?. Surg Oncol, 17:253–258.

21.

Jeong

, Baek

, Son

, Choi

, Kim

, Ko

, Chung

, Baek

. 2006. Integrated 18F-FDG PET/CT for the initial evaluation of cervical node level of patients with papillary thyroid carcinoma: comparison with ultrasound and contrast-enhanced CT. Clin Endocrinol (Oxf), 65:402–407.

22.

Gemsenjager

, Perren

, Seifert

, Schuler

, Schweizer

, Heitz

. 2003. Lymph node surgery in papillary thyroid carcinoma. J Am Coll Surg, 197:182–190.

23.

Bardet

, Malville

, Rame

, Babin

, Samama

, De Raucourt

, Michels

, Reznik

, Henry-Amar

. 2008. Macroscopic lymph-node involvement and neck dissection predict lymph-node recurrence in papillary thyroid carcinoma. Eur J Endocrinol, 158:551–560.

24.

Ito

, Tomoda

, Uruno

, Takamura

, Miya

, Kobayashi

, Matsuzuka

, Kuma

, Miyauchi

. 2004. Preoperative ultrasonographic examination for lymph node metastasis: usefulness when designing lymph node dissection for papillary microcarcinoma of the thyroid. World J Surg, 28:498–501.

25.

Loevner

, Kaplan

, Cunnane

, Moonis

. 2008. Cross-sectional imaging of the thyroid gland. Neuroimaging Clin N Am, 18:445–461, vii.