| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Università "La Sapienza" 324-00161 Rome, Italy
In their recently published article Moritz et al. [1] effectively showed the clinical usefulness of contrast-enhanced color Doppler sonography in the differential diagnosis of enlarged cervical lymph nodes. I agree with the authors that semi-quantitative indexes of flow (such as resistive index and pulsatility index) cannot be effectively proposed to differentiate benign from malignant nodes, mostly because it is nearly impossible to obtain repeatable measurements in smaller nodes.
When the authors discuss vascular patterns of enlarged lymph nodes resulting from both unenhanced and contrast-enhanced color Doppler sonography, they affirm that "reactively enlarged nodes characteristically showed hilar vessels branching in the center..." and that "in lymph node metastases, vessels were found predominantly in the periphery...." On the basis of these observations they claim that "the resulting sensitivity and specificity were 100% and 98% for lymph node metastases and 98% and 100% for reactively enlarged lymph nodes." The authors included in the references one of the articles I published on this topic in 1997 [2], which is cited in the text as follows: "a similar vascular pattern has been described by Giovagnorio et al.; however, a differentiation between acute inflammation and metastasis was apparently not attempted in that study."
Unfortunately the authors do not seem to have read my article carefully because they did not notice that it introduced the same distinction between hilar and peripheral vascular patterns that they redescribed and that this distinction is the base on which a differential diagnosis between inflammation and metastasis was attempted. Moreover, the hilar pattern was further subdivided into two subpatterns, a "normal hilar" (type I, with evidence of a single vascular pole with possible visibility of small regular branches) and a "hypertrophic hilar" (type II, with the main hilar feeding artery almost doubled in diameter and length, with evidence of two or more regular branches), whereas the "peripheral" pattern was classified as "mainly peripheral vascularity, with three or more vascular branches perforating the capsule peripherally and directed toward the center of the node" (type III). The "normal hilar" pattern was associated with chronic inflammation with a sensitivity of 85% and a specificity of 90%, the "hypertrophic hilar" was associated with acute inflammation with a sensitivity of 68% and a specificity of 55%, and the "peripheral" pattern was associated with metastasis with a sensitivity of 47% and a specificity of 91%.
I believe that the administration of contrast medium should increase the sensitivity and specificity of the associations between vascular patterns and specific diseases, mostly because contrast-enhanced color Doppler sonography can reveal slower flow in smaller arterial branches encased by neoplastic tissue or compressed by fibrosis. This modality, therefore, can help avoid the misdiagnosis of inflammation in a falsely avascular or hypovascular node. This consideration certainly explains why the authors obtained higher specificity and sensitivity, but they should have pointed out that the association between certain vascular patterns and specific diseases had already been documented in 1997.
References
Justus-Liebig-Universität Giessen 35392 Giessen, Germany
We described in our article the possibility of differentiating reactively enlarged lymph nodes and lymph node metastasis on the basis of their vascular pattern [1]. Hilar vessels were found in reactively enlarged lymph nodes, whereas vessels predominantly in the periphery of the nodes were most often seen in metastasis. We clearly noted "a similar vascular pattern has been described by Giovagnorio et al." [1]. The statement "a differentiation between acute inflammation and metastasis was apparently not attempted in that study" was based on the sentence "the second and third" type of vascular patterns "were related, to a lesser extent, to acute inflammation and neoplasm (sensitivity 68% and 55%, specificity 47% and 91%, respectively)" [2]. We are sorry to misinterpret this sentence.
We agree absolutely with Giovagnorio in distinguishing three groups of vascular patterns in lymph nodes. We examined a select group of patients, all with carcinomas of the oral cavity or the immediate vicinity. It did not seem useful to differentiate between chronic and acute inflammation in this clientele because most patients with reactively enlarged lymph nodes suffered from chronic inflammation. Our experience in children with enlarged lymph nodes, however, shows that a classification of the vascular pattern into three groups might be useful, although a considerable overlap is observed. For example chronic inflammation due to atypical mycobacteriosis showed a type II vascular pattern with a main hilar feeding artery and two or more regular branches. Acute inflammation with mononucleosis resulted in some cases in a type III vascular pattern with mainly peripheral vascularity. On the other hand in cases of lymphoma, a type II vascular pattern was found. Larger studies are needed to show the sensitivity and specificity of the vascular pattern as a single parameter in differentiating between chronic inflammation, acute inflammation, and malignant lymph node enlargement. In combination with the clinical signs, however, the vascular pattern seems to be a sensitive parameter for differentiating benign and malignant lymph node enlargement.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
