| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology, University of Pittsburgh Medical Center, 200 Lothrop
St., PUH Rm. D132, Pittsburgh, PA 15213.
2 Department of Otolaryngology, University of Pittsburgh Medical Center,
Pittsburgh, PA 15213.
Received July 7, 2005;
accepted after revision October 21, 2005.
Address correspondence to B. F. Branstetter
(bfb1{at}pitt.edu).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. In a review of electronic medical records from January 2001 to January 2003 we identified the cases of 500 patients who had undergone CT of the neck. Because they had confounding factors such as cancer or cervical lymphadenopathy, 199 of these patients were excluded, leaving 301 patients in the study: 131 (44%) male patients and 170 (56%) female patients. The age range was 11-92 years (average age, 46 years). Two head and neck radiologists used strict diagnostic criteria and consensus to identify the distal thoracic duct on both sides of the neck. One half of the images selected at random were flipped left to right. The purpose of randomization was to avoid interpretation bias, because the thoracic duct is known to typically course within the left side of the neck. The configuration of the distal duct was tabulated, and effects of age and sex were statistically evaluated.
RESULTS. The left side of the neck was unevaluable in 26 (9%) of 301 patients because of streak artifact. In the other 275 patients, the distal thoracic duct was identified in the left side of the necks of 150 (55%) of the patients. Eleven of these patients (4%) also had a visible duct in the right side of the neck, but a right-sided duct was never identified without a left-sided counterpart. The distal thoracic duct had a tubular configuration in 70 (43%), a flared configuration in 72 (45%), and a long segmental fusiform dilation in 19 (12%) of 161 patients. Patient sex had no significant effect on the appearance of the distal thoracic duct. Older patient age had a slight positive effect on the size of the duct.
CONCLUSION. Familiarity with the normal CT appearance of the distal thoracic duct can be helpful in differentiating a normal duct from pathologic lesions of the lower neck, such as lymphadenopathy.
Keywords: CT • cystic neck mass • lymph nodes • neck • thoracic duct
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
On radiographic images, distal thoracic duct enlargement can be mistaken for a reactive lymph node, a congenital cyst, or nodal metastasis. Knowledge of the normal CT anatomy of the cervical thoracic duct is important to avoid mistaking a dilated distal duct for a lesion on cross-sectional imaging and to avoid unnecessary intervention. The purpose of this study was to determine how often the distal thoracic duct can be identified on neck CT and to characterize its appearance.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
Image Interpretation
All CT scans were obtained with a HiSpeed or a LightSpeed scanner (GE
Healthcare) at a slice thickness of 2.5 mm or 3.0 mm after administration of
125 mL of nonionic IV contrast material (Optiray, Mallinckrodt). Consensus
interpretation was rendered by two head and neck radiologists. Both were
fellowship trained and possessed certificates of added qualification in
neuroradiology, and both spend most of their clinical time on head and neck
imaging. Images were viewed on a PACS workstation (Stentor) in axial stack
mode without reconstructions. All measurements were obtained with the
measurement tools inherent to the PACS workstation. The following strict
criteria were required for classification of a lower neck structure as the
distal thoracic duct: tubular structure 3 2 mm in diameter in the
lower neck; termination into a vein at or near the confluence of the internal
jugular and subclavian veins; posteromedial extension behind the common
carotid artery toward the tracheoesophageal groove; inferior continuation near
the tracheoesophageal groove over at least two contiguous images; distinct
appearance from the vertebral and suprascapular veins. Even if both
interpreting radiologists were convinced of the presence of a visible thoracic
duct, the duct was tabulated only when all criteria were met.
One half of the images selected at random were flipped left to right during image interpretation. The two radiologists were blinded to this flipping so that they would not be more inclined to conclude when evaluating the left side of a patient that a thoracic duct was present. This step also served as an internal control for determining whether the criteria for identifying the thoracic duct were overly lax (identification of many ducts on the right side would indicate that the criteria were too lax because structures other than the thoracic duct were being classified as ducts).
IV contrast material was administered to each patient through an antecubital vein in either the left or the right arm. In some patients, streak artifact from an undiluted contrast bolus in the subclavian vein prevented evaluation of one side of the neck. The affected side of the neck was excluded from statistical analysis, but the evaluable side of the neck remained in the study. After application of this exclusion criterion, 275 left sides of necks and 288 right sides of necks remained in the study.
|
|
Statistical Analysis
Statistical analyses of distal left thoracic duct size and frequency of
left thoracic duct identification were performed with the 275 cases in which
the left side of the neck was evaluable. Patient age and sex were used as
independent variables. The right sides of the neck were not independently
analyzed because too few thoracic ducts were identified on the right. To
analyze sex, we used a chi-square test to assess frequency of identification
and a Student's t test to assess the average size of the duct when it
was seen. To analyze the effect of patient age, we used a Student's t
test to assess frequency of identification and applied an F test to a linear
regression model to assess the effect of age on duct size. We intentionally
chose a consensus model for our analysis rather than comparing the
interpretations of the two observers using kappa statistics because the
purpose of this study was to describe the appearance of an entity, not to test
the precision of the technique for identification of that entity.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Duct Configuration
Distal duct configuration was assessed on all of the identified thoracic
ducts (n = 161). Simple tubular shape, whereby the duct maintained
the same diameter to the point of termination
(Fig. 1), was seen in 70 (43%)
of the 161 cases. Flared configuration, whereby the distal duct flared out
just before emptying into the venous confluence
(Fig. 2), was seen in 72 (45%)
of the 161 cases. Segmental fusiform configuration, representing segmental
dilation of the distal duct (Fig.
3), was seen in 19 (12%) of 161 cases.
Patient Sex
In male patients, the duct was identified in 70 (57%) of 123 left sides of
the neck. In female patients, the duct was identified in 80 (53%) of 152 left
sides of the neck. This difference was not statistically significant
(p = 0.48). In male patients, the average duct diameter was 4.67 mm;
in female patients, the average diameter was 4.85 mm. This difference also was
not significant (p = 0.60).
Patient Age
In the 275 patients in whom the left side of the neck was evaluable, the
average age and age range were the same as those of the total set of 301
patients. The average age of patients with a visible distal thoracic duct was
45.5 years; the average age of patients without a visible duct was 46.6 years
(p = 0.65). There was a statistically significant association between
greater diameter of identified distal ducts and older patient age (F observed
value = 4.83; F critical value = 3.91). The association was 0.02 mm of duct
diameter per year of life.
|
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Embryologically, the thoracic duct is formed from six lymph sacs. The paired jugular lymph sacs appear near the junction of the internal jugular veins and the subclavian veins [21]. The retroperitoneal lymph sac develops near the primitive inferior vena cava, and the cisterna chyli develops dorsal to the aorta. The paired posterior lymph sacs form near the junction of the primitive iliac veins. Linkage of the cisterna chyli with the jugular lymph sacs early in fetal life produces a bilateral system of lymphatic trunks [22]. The definitive thoracic duct is then formed by the cisterna chyli, the lower two thirds of the right lymphatic trunk, and the upper third of the left lymphatic trunk. The thoracic duct has a typical course in 60-70% of persons [22]. In a typical duct, the cisterna chyli is found at the level of the first lumbar vertebra on the right, and the duct then ascends through the right abdomen and chest. At the level of the fourth thoracic vertebra, the thoracic duct crosses to the left and continues superiorly, forming an arch that rises 3-5 cm above the clavicle [21, 23]. The duct then angles forward, running posterior to the left common carotid artery, and terminates in the posterolateral aspect of the venous angle (the junction of the left internal jugular and subclavian veins). Morphologic studies of cadavers have shown that the normal distal duct can be tubular or dilated in an ampullar (flared) form [23-25]. Other points of termination are the left internal jugular vein, left external jugular vein, brachiocephalic vein, and left subclavian vein. Because the lymphatic trunks are originally bilateral, several variations of the final course of the duct have been described, including bilateral thoracic ducts and multiple distal branches.
This study showed that the cervical thoracic duct frequently can be identified on CT. Knowledge of its normal course (Figs. 4A and 4B) and typical appearance (Figs. 1, 2, 3) is essential in differentiating the duct from pathologic conditions. However, there are several potential mimics of a dilated distal duct. Most important is Vir- chow's lymph node, enlargement of which may be the first sign of a malignant lesion arising in the chest or abdomen (Fig. 5). Nearby veins, namely the external jugular, anterior jugular, suprascapular, and vertebral veins, can be confusing tubular structures in the region of the distal duct. One example is a dilated vertebral vein (Fig. 6). The vertebral vein can be differentiated on the basis of its position (lateral to the expected course of the distal thoracic duct) and course (superior into the neck rather than inferior into the chest like that of the duct). Another mimic is unopacified blood from the extremity mixing with opacified blood from the neck during IV contrast administration (Fig. 7). The mixed blood can simulate a flared distal duct. In equivocal cases, MRI (Figs. 8A and 8B) or sonography can be useful. Sonographic characteristics, including flow rates, have not yet been defined in the literature.
|
|
Another potential differentiating factor is density of the distal duct. In theory, chyle in the thoracic duct should be of lower attenuation than blood in veins and fluid in cysts. Unfortunately, attenuation measurements on 2-mm structures are technically challenging, especially in imaging of the lower part of the neck, in which beam hardening and streak artifacts are prevalent. Thus density is not considered a practical distinguishing characteristic.
Our results showed that the cervical thoracic duct was identified on CT in the left side of the neck in 55% of subjects in our series. We found no right-sided ducts in isolation, despite having blinded the radiologists to left- and right-sidedness. This finding suggests that our criteria for identification of the duct are sufficiently specific and that the observers were not biased by the side of the neck they were evaluating.
A structure was identified as the distal thoracic duct only when all the specified criteria were met. We excluded many structures that likely were the distal duct but did not strictly meet the criteria. Thus our results likely are an underestimate of the actual percentage of patients in whom the distal thoracic duct is visible, which may be as high as 75% on the basis of the subjective impressions of our observers.
Older patient age was associated with a slightly greater diameter of identified distal thoracic ducts. Although it was statistically significant because of the large number of patients in the study, this effect amounted to only 0.02 mm per year of life, so the effect is unlikely to be appreciable clinically or radiographically. Sex was not a significant predictor of the appearance of the distal thoracic duct.
Our study had several limitations. Although efforts were made to control for bias by flipping half of the studies left-to-right, sidedness often can be deciphered by the position of the cervical vessels or esophagus. No surgical confirmation of the nature of the imaging findings was obtained because the structures are considered normal. Instead, confirmation of the results was based on knowledge of the normal anatomic course of the thoracic duct.
Important features helpful in identifying the distal thoracic duct on CT are characteristic location near the junction of internal jugular and subclavian veins, typical sigmoid course toward the tracheoesophageal groove and then down into the chest, and tubular configuration. Familiarity with the normal CT appearance of the distal thoracic duct can be helpful in differentiating a normal duct from a lesion of the lower neck.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  G. N. Matwiyoff, D. A. Bradshaw, K. H. Hildebrandt, J. F. Campenot, J. M. Coletta, and W. J. Coyle A 28-Year-Old Man With a Mediastinal Mass Chest, September 1, 2008; 134(3): 648 - 652. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
