| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Radiology, Shinshu University School of Medicine, 3-1-1 Asahi,
Matsumoto 390-8621, Japan.
2 Department of Surgery, Shinshu University School of Medicine, Matsumoto
390-8621, Japan.
Received May 28, 2002;
accepted after revision August 6, 2002.
Address correspondence to S. Takashima.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. We reviewed MR imaging findings (lesion size and posterior extension of tumor; encirclement of and invasion into the laryngeal cartilage, trachea, and esophagus by the tumor; and the amount of effaced fatty tissue in the tracheoesophageal groove or between the laryngeal cartilage and hypopharyngeal wall) in 66 patients with thyroid carcinoma. The amount of effaced fatty tissue was classified as one of five grades: 1, normal amount of fatty tissue; 2, partly effaced; 3, completely effaced in one MR imaging slice; 4, completely effaced in two contiguous MR slices; and 5, completely effaced in three or more contiguous MR slices.
RESULTS. Thirty-two (48%) of the 66 patients had surgically or pathologically verified recurrent laryngeal nerve invasion. Logistic modeling revealed that the amount of effaced fatty tissue (p < 0.001) and the lesion size (p = 0.033) were the significant factors. Using the threshold values for the lesion size to predict invasion, we found that a threshold of more than 2.9 cm showed the highest accuracy, 76%, with 78% sensitivity and 74% specificity. For the amount of effaced fatty tissue, a grade of 3 or more had the highest accuracy, 88%, with 94% sensitivity and 82% specificity. Addition of the lesion size to this criterion did not improve the diagnostic accuracy of using the amount of effaced fatty tissue alone.
CONCLUSION. Invasion of the recurrent laryngeal nerve by thyroid carcinoma was accurately predicted by the finding of effaced fatty tissue on MR imaging.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although unilateral recurrent laryngeal nerve paralysis can be caused by various lesions, thyroid carcinoma is the most common cause of the paralysis if idiopathic origins are excluded [5, 6]. Hoie et al. [7] documented an 11% prevalence of the recurrent laryngeal nerve invasion in 1245 patients with thyroid carcinomas. However, when the study population was limited to patients with locally invasive thyroid carcinomas, the prevalence rose to as much as 47% [3]. Three other studies described imaging analyses of tumor invasion by thyroid carcinoma to the regional lymph nodes, trachea, and esophagus [8,9,10]. In addition, three studies described morphologic changes in laryngeal and pharyngeal structures caused by paralysis of the recurrent laryngeal nerve [2, 6, 11]. However, to our knowledge, recurrent laryngeal nerve invasion by thyroid carcinoma has rarely been described in the radiology literature. In this study, we investigated MR imaging features of recurrent laryngeal nerve invasion by locally invasive thyroid carcinomas and defined an optimal MR imaging finding for predicting recurrent laryngeal nerve invasion by the tumors.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MR Imaging Techniques
MR imaging was performed with a 1.5-T unit (Magnetom, Siemens, Erlangen,
Germany; or Signa, General Electric Medical Systems, Milwaukee, WI) using a
volume neck coil or a Helmholtz coil. T1-weighted MR imaging (for 29 patients:
TR/TE, 700/13; excitations 2; and for 37 patients: 900/15; excitation, 1) and
conventional spin-echo T2-weighted MR imaging (for 37 patients: 2000/70;
excitation, 1) or fast spin-echo T2-weighted MR imaging (for 29 patients:
3200/91; excitations, 2) were performed in the transverse plane. From the
level of the mandibular angle to the sternal notch, 18 images were obtained
with a slice thickness of 5 mm and an intersection gap of 1-2 mm. The field of
view was 20 x 20 cm, and the acquisition matrix was 256 x 192.
After IV bolus injection of 0.2 mmol/kg of gadopentetate dimeglumine
(Magnevist; Schering, Berlin, Germany), contrast-enhanced transverse
T1-weighted MR imaging (for 29 patients: 700/13; excitations, 2; and for 25
patients: 900/15; excitation, 1) without fat suppression was performed in 54
(82%) of the 66 patients.
MR Imaging Analysis and Statistical Methods
Two radiologists independently evaluated the MR images of the 66 patients
with thyroid carcinoma without knowledge of surgical and pathologic findings.
First, the two radiologists determined whether the vocal cord was paralyzed on
the basis of imaging findings such as thyroarytenoid atrophy, anteromedial
deviation of the arytenoid cartilage, enlarged piriform sinus, enlarged
laryngeal ventricle, paramedian cord, or atrophy of the posterior
cricoarytenoid muscle [6,
11]. Then, they assessed MR
findings with regard to the maximal transverse diameter of thyroid tumors
(lesion size) and the presence of tumor extension with poorly defined margins
beyond the posterior portions of the thyroid gland (posterior extension);
tumor invasion to the esophageal wall, trachea, or laryngeal cartilage below
the level of the cricothyroid joint; tumor encirclement of the esophagus,
trachea, or cartilage; and amount of effacement of fatty tissue in the
tracheoesophageal groove or between the hypopharyngeal wall and the laryngeal
cartilage below the level of the crycothyroid joint. In patients in whom the
tumor was present in both lobes of the thyroid gland, the more severely
affected side was used for evaluation.
The degree of tumor encirclement of each organ was classified as one of five grades: grade 1, no encirclement; grade 2, 1-90°; grade 3, 91-180°; grade 4, 181-270°; and grade 5, 271-360°. The amount of effaced fatty tissue was also classified into four grades: grade 1, normal amount of fatty tissue; grade 2, partly effaced; grade 3, completely effaced on one MR slice; grade 4, completely effaced on two contiguous MR slices; and grade 5, completely effaced on three or more contiguous MR slices. The criteria of tumor invasion were the presence of focally increased T2 signal in the esophageal wall for the esophagus [9], soft tissue in the tracheal cartilage for the trachea [10], and destruction of the cartilage by tumor or tumor wrapping around the posterior edge of the thyroid cartilage for the laryngeal cartilage [3].
Interobserver agreement between the two reviewers was calculated using kappa statistics for data in eight categories (posterior tumor extension; tumor invasion to the esophagus, trachea, and cartilage; tumor encirclement of the esophagus, trachea, and cartilage; and amount of effaced fatty tissue) and using correlation coefficients for lesion size. If interpretations for the data in a category differed, a third radiologist reviewed the cases, and the majority opinion became the final decision. The averaged values of the two measurements were used for analysis of lesion size.
Stepwise logistic modeling was conducted using all nine parameters as independent variables to determine a significant factor. We proposed an optimal criterion for recurrent laryngeal nerve invasion. The correlation of lesion size was examined using Pearson's correlation coefficient. Unpaired Student's t tests were used for comparison of lesion size. Fisher's exact tests were used to compare the prevalence of each MR finding. A p value of less than 0.05 was considered statistically significant. We used a statistical software package (SPSS, Chicago, IL) for our calculations.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The affected recurrent laryngeal nerve and the thyroid tumor were resected together in 26 (81%) of the 32 patients with confirmed tumor invasion, and the nerve was left unresected in the residual tumor in six patients (19%). Pathologic proof of tumor invasion to the affected recurrent laryngeal nerve was obtained in six (23%) of the 26 patients with resected recurrent laryngeal nerves. In the other 20 patients (77%), resected recurrent laryngeal nerves were not identifiable in pathologic specimens. Of the 32 patients with recurrent laryngeal nerve invasion, the nerve was invaded by primary thyroid tumor in 30 patients (94%) and by metastatic tumor in two patients (6%).
Correlation between surgical and MR findings revealed that tumor-infiltrated portions of the recurrent laryngeal nerve had been included in the areas scanned by MR imaging in all 32 patients. Of the 32 patients with recurrent laryngeal nerve invasion, 30 patients (94%) had surgically or pathologically confirmed tumor invasion to the other organs such as the trachea (n = 18; 56%), esophagus (n = 16; 50%), internal jugular vein (n = 15; 47%), laryngeal cartilage (n = 12; 38%), common carotid artery (n = 8; 25%), or regional lymph nodes (n = 26; 81%). The remaining two patients (6%) had tumor invasion only to the recurrent laryngeal nerve. Tumor invasion to the vagus nerves by metastatic lymph nodes was coexistent in three (9%) of the 32 patients.
MR Imaging Findings
The kappa value of the data from the eight categories ranged from 0.43 to
0.83, with a mean value of 0.68, indicating moderate to almost perfect
agreement [12]. Pearson's
correlation coefficient for lesion size was 0.89 (p < 0.001),
indicating strong agreement of the two measurements.
As summarized in Table 1, lesion size was statistically significantly larger (p < 0.001) in patients with invasion of the recurrent laryngeal nerve (3.8 ± 1.3 cm) than in those without invasion (2.7 ± 1.7 cm). The prevalence of posterior tumor extension (p < 0.001), esophageal invasion (p = 0.003), tracheal invasion (p = 0.001), and cartilage invasion (p = 0.001) was significantly greater in patients with recurrent laryngeal nerve invasion than in patients without recurrent laryngeal nerve invasion. Significant differences between patients with and those without recurrent laryngeal nerve invasion were found for the tumor encirclement of the esophagus (p = 0.002), trachea (p = 0.004), and cartilage (p = 0.001), and for the amount of effaced fatty tissue (p < 0.001).
|
Diagnostic Statistics
Diagnostic statistics for each MR feature and morphologic assessment are
summarized in Table 2.
Morphologic assessment had an accuracy of 68% and a high specificity of 91%,
but its sensitivity was low (44%) (Fig.
1A,1B,1C,1D).
Of the various threshold values of lesion size, the value of larger than 2.9
cm had the highest accuracy of 76%, with 78% sensitivity and 74% specificity.
For the degree of tumor encirclement of the esophagus, a grade of 2 or more
showed the highest accuracy of 73%, with 91% sensitivity and 56% specificity.
For the degree of tumor encirclement of the trachea, a grade of 3 or more had
the highest accuracy (66%), with 76% sensitivity and 56% specificity. For the
degree of tumor encirclement of the cartilage, a grade of 3 or more showed the
highest accuracy of 68%, with 91% sensitivity and 47% specificity. For the
amount of effaced fatty tissue, a grade of 3 or more had the highest accuracy
(88%), with 94% sensitivity and 82% specificity. A grade of 2 or more achieved
100% sensitivity, but its specificity was 53%. Of the four binary MR imaging
criteria—the presence or absence of posterior extension of the tumor or
tumor invasion into the esophagus, trachea, and cartilage—posterior
tumor extension had the highest accuracy (73%), with 53% sensitivity and 91%
specificity.
|
|
|
|
|
In selecting one single MR imaging criterion with which to predict invasion of a recurrent laryngeal nerve by tumor, we found the highest accuracy (88%) was achieved using the amount of effaced fatty tissue (Fig. 1A,1B,1C,1D). Use of this criterion produced two false-negative and six false-positive diagnoses. One patient with papillary thyroid carcinoma given a false-negative diagnosis had lymph node metastases in the lower neck that invaded the recurrent laryngeal nerve, and another patient with a false-negative diagnosis had anaplastic thyroid carcinoma (Figs. 2A,2B and 3). Of the six patients with a false-positive diagnosis, five had a primary thyroid tumor larger than 2.9 cm.
|
|
|
The stepwise logistic modeling revealed that the amount of effaced fatty tissue (p < 0.001) and size of the lesion (p = 0.033) were the significant factors for predicting recurrent laryngeal nerve invasion. However, addition of the factor of lesion size to the factor of effaced fatty tissue did not increase accuracy above that of the effaced fatty tissue alone (Table 2).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Controversy exists regarding the amount of the tumor to remove if the tumor is invading the surrounding organs. Some surgeons insist on complete en bloc resection of the tumor to obtain clear margins [15,16,17], whereas others recommend shaving the tumor, although microscopic disease may be left behind [3, 18, 19]. Nonetheless, the general consensus is that the resection of the entire gross tumor is vital to prevent local recurrence and improve the chances of survival. Because locally advanced thyroid tumors frequently infiltrate not only the recurrent laryngeal nerve but also other adjacent aerodigestive organs, tumor invasion to only the recurrent laryngeal nerve is rare [3, 19]. Actually, most of our patients (94%) with invasion of the recurrent laryngeal nerve had coexistent tumor invasion to the aerodigestive structures or major vessels. Therefore, the involved recurrent laryngeal nerve—regardless of whether it was functioning or not—was excised along with the thyroid tumors to remove all macroscopic tumors.
The sacrifice of functioning recurrent laryngeal nerves reduces the quality of the patient's life by impairing the airway, voice, and swallowing [3, 4]. In our study, 20 of the 26 patients who underwent resection of the recurrent laryngeal nerve and thyroid tumor had no pathologic verification of tumor invasion; we considered these patients to have had a surgical diagnosis of tumor invasion because that is standard practice in many institutions. In our series, we found that laryngoscopic findings and clinical symptoms were not reliable tools for predicting tumor invasion of the recurrent laryngeal nerve. Only 38% of the patients with tumor invasion to the recurrent laryngeal nerve developed clinical symptoms, and only 69% of the patients had abnormal laryngoscopic findings. Several articles have shown that tumor infiltration of the recurrent laryngeal nerve does not necessarily result in vocal cord paralysis and that even benign conditions can cause vocal cord paralysis [13, 14]. In one series, 50 (51%) of 98 patients with invasion into the surrounding tissues had invasion into the recurrent laryngeal nerve but laryngoscopic findings showed them to have intact normal vocal cord function [13]. Kowalski and Filho [19] found that only two (13%) of their 15 patients with recurrent laryngeal nerve invasion showed clinical symptoms. Because clinical symptoms and laryngoscopic findings are not diagnostic of recurrent laryngeal nerve invasion, establishing a reliable sign on imaging findings of such an invasion would be clinically beneficial.
In our analysis, diagnosis using morphologic changes in the laryngeal or pharyngeal structures on MR imaging showed only 44% sensitivity. Romo and Curtin [6] reported that the paramedian cord and enlarged piriform sinus were most frequently observed in patients with recurrent laryngeal nerve palsy and that all of their 20 patients exhibited both of these features on CT or MR imaging or both. However, all patients in that series had a history of vocal cord paralysis, and furthermore nearly all patients were evaluated using CT scans obtained with a helical technique (a thinner section thickness of 2-5 mm with a table speed of 3-5 mm/sec). On the other hand, the percentage of patients with clinically proven vocal cord paralysis in our series was low, and we evaluated MR images obtained using a thicker section thickness (5 mm) with intersection gaps (1-2 mm) and a much longer acquisition time than that used in helical CT.
According to the logistic modeling used in our study, the amount of effaced fatty tissue and the size of the lesion were significant factors for predicting invasion of the recurrent laryngeal nerve by the thyroid tumor. However, using lesion size and the amount of the effaced fatty tissue together as predictors did not result in greater accuracy than using the amount of effaced fatty tissue alone. Of all the criteria, effaced fatty tissue on one or more contiguous MR slices was the factor with the highest accuracy, 88%, with 94% sensitivity and 82% specificity. We adopted this MR imaging feature as a potential criterion because most of the track of the recurrent laryngeal nerves lies in the fatty tissue in the tracheoesophageal groove and between the tracheal and hypopharyngeal wall and because we thought that the fatty tissue in those areas might be obliterated if a thyroid tumor invaded the recurrent laryngeal nerves [1, 20]. We believe that our MR imaging criterion of effaced fatty tissue can be applied to CT findings because CT can reveal the fatty tissue in the tracheoesophageal groove as well as MR imaging.
This criterion produced two false-negative diagnoses, which were due to anaplastic carcinoma in one patient and invasion by a metastatic lymph node in the lower neck in another. Infiltration of the recurrent laryngeal nerve by metastatic nodes in this region has previously been documented in the literature [21]. Normal recurrent laryngeal nerves usually pass 1-2 cm lateral relative to the trachea in the region of the lower pole of the thyroid gland [22]. Therefore, our MR imaging criterion may be less accurate for assessment of invasion in the level below the entrance of the nerve into the tracheoesophageal groove. Several variations in the course of the recurrent laryngeal nerves have been reported, and a variant that branches from the vagus nerve in the neck (nonrecurrent laryngeal nerves) is reported to occur in 0.52% of patients [23, 24]. Such anomalies can result in a false-negative diagnosis. Additionally, tumor invasion to the vagus nerve but not to the recurrent laryngeal nerve can also lead to a false-negative diagnosis because the vagus nerve descends into the carotid space in the neck [1, 2].
The false-positive diagnoses in our study occurred mostly in patients with large thyroid tumors. Although patients with large thyroid nodules and benign thyroid goiters were not included in our series, these entities may mimic tumor invasion of the recurrent laryngeal nerve because of the obliteration of fatty tissue. Thus, we suggest that our criterion for predicting tumor invasion is only valid for well-differentiated carcinomas or anaplastic carcinomas in the thyroid gland and should be used at the level of the thyroid gland.
In our study, we used MR images obtained using standard techniques only; therefore, we were able to investigate only indirect MR imaging features in predicting invasion of the recurrent laryngeal nerve because depiction of the nerve itself is impossible using standard MR imaging techniques. With recent advances in MR imaging technology, direct visualization of many peripheral nerves is now possible with MR neurography using high-resolution T1-weighted MR imaging sequences and fat-suppressed T2-weighted fast spin-echo or short tau inversion recovery MR imaging sequences in conjunction with a phased array radiofrequency coil [25, 26]. These techniques can reveal the presence and extent of neural tumors, delineate nerve continuity, and depict enlargement of and abnormal signal intensity in diseased peripheral nerves. Therefore, in the future, MR neurography may improve the diagnostic accuracy of invasion of the recurrent laryngeal nerve by adjacent tumors.
In conclusion, our new MR imaging criterion was accurate for predicting invasion of the recurrent laryngeal nerve and will be clinically useful for preoperatively selecting patients who should undergo excision of functioning recurrent laryngeal nerves.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  N. A. Johnson and M. E. Tublin Postoperative Surveillance of Differentiated Thyroid Carcinoma: Rationale, Techniques, and Controversies Radiology, November 1, 2008; 249(2): 429 - 444. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. G. Lohan, A. Tomasian, R. Saleh, M. Krishnam, and J. P. Finn Hypervascular Thyroid Nodules on Time-Resolved MR Angiography at 3 T: Radiologic-Pathologic Correlation Am. J. Roentgenol., April 1, 2008; 190(4): W255 - W260. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
