| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Report |
1
Department of Radiology, University of Vienna, Waehringer Guertel 18-20,
A-1090 Vienna, Austria.
2
Department of Pediatric Surgery, University of Vienna, A-1090 Vienna,
Austria.
Received May 10, 1999;
accepted after revision August 9, 1999.
Presented at the annual meeting of the American Roentgen Ray Society,
Boston, May 1997.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CONCLUSION. MR imaging and CT may substantially broaden visualization of the spectrum of abnormalities seen in generalized cystic lymphangiomatosis by revealing the complete extent of disease and, thus, may contribute to clinical management of the disease by preventing initial misdiagnosis.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In an individual patient, the spectrum of conventional radiologic findings displayed by generalized cystic lymphangiomatosis may eventually be suggestive of the disease [2,3,4,5,6]. Because of the rarity of generalized cystic lymphangiomatosis, however, reports on the overall spectrum of imaging findings seen in this disease are sparse. In part, these findings were documented with imaging techniques, such as lymphography, that are no longer used [7]. Although the use of sonography, CT, and MR imaging has been described, these descriptions were often limited to a single anatomic compartment, such as the mediastinum, the retroperitoneal space, or the peritoneal cavity [2, 3, 6]. The aim of our investigation, therefore, was to illustrate a wider radiologic spectrum of generalized cystic lymphangiomatosis imaged by CT and MR imaging and compare our findings with gross-pathologic specimens.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
Diagnostic Imaging
All CT studies were performed on a Somatom Plus S scanner (Siemens Medical
Systems, Erlangen, Germany) with standard protocols used at our institution.
Cranial CT studies were performed without contrast material, using slice
thicknesses of 3 and 5 mm. Images were photographed with window settings
appropriate for visualization of bones, soft tissues, and lung parenchyma.
Thoracic and abdominal CT were performed in the helical mode with a slice
thickness of 8 mm, a table increment of 6 mm, and a reconstruction index of 3
mm after IV administration of 1 ml nonionated contrast material per kilogram
of body weight.
All MR studies were performed on a Magnetom 1.5-T unit (Siemens Medical Systems) according to standard protocols used at our institution. T1 - and T2-weighted turbo spin-echo sequences, T2-weighted inversion-recovery sequences, and fat-saturated sequences were acquired in axial, coronal, and sagittal planes. For MR studies of the brain, additional T1-weighted gradient-echo for multiplanar reconstructions and fluid-attenuated inversion-recovery sequences were performed. Slice thickness was adapted individually according to the anatomic region of interest with respect to the body size of each patient.
Image Analysis
All images were analyzed independently by two radiologists experienced in
pediatric imaging with special interest to the extent and size of generalized
cystic lymphangiomatosis—associated lesions, organ involvement, and CT
attenuation and MR signal characteristics of the depicted lesions. Reviewers
were aware of clinical histories, histopathology, and surgical reports.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
|
|
|
Pleural effusions were seen in all four patients examined by CT and in three patients examined by MR imaging. Cytologic workup of these pleural effusions revealed lymphangitic chyle.
Pulmonary interstitial nodules were present in all four patients examined with MR imaging and in three patients examined with CT. Histopathology showed chyle-filled cystic tumors along the interstitial lymphatic pathways.
Less common findings included cystic lesions of the spleen (Fig. 5), seen in three patients with CT and in two patients with MR imaging, and nodular thickening of the bowel wall, seen in one child examined with MR imaging.
|
Two patients also showed subcutaneous nodular and partly cystic lesions with surrounding edema. In both patients, these lesions were originally mistaken for nonspecific inflammatory alterations.
On both CT and MR imaging, the depicted lesions did not enhance after administration of contrast material. All lesions were sharply delineated, and the wall of the cystic tumors was always visible. Pathologic bone fractures of different age and different healing status were seen in all patients, but none of the patients had any periostal reactions or adjacent tumors. Distribution of abnormalities was diffuse in all patients, and none of the patients had lesions confined to a single organ. None of the lesions had rim or coarse calcifications. Distribution of the lesions also was diffuse within the organs and showed neither central nor peripheral predominance. The size of the individual lesions ranged from 2 mm to 13 cm. Histopathologic workup of parenchymal and osseous lesions revealed chyle-filled multiseptated cystic tumors—that is, simple and cavernous lymphangiomas and cystic hygromas (depending on the histopathologic classification used by the pathologists).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The first of these finding was extensive mediastinal involvement, as seen in four of our five patients. Interestingly, all four of these patients had simultaneous intrapulmonary involvement and pleural effusions. Brown et al. [11] and Pilla et al. [3] found that mediastinal involvement in generalized cystic lymphangiomatosis occurs only in adults. On the basis of our results, we cannot confirm this statement. The same researchers, however, also described pleural effusions associated with mediastinal involvement in adult patients with generalized cystic lymphangiomatosis [3, 11]. This observation was confirmed in our pediatric patients. We may speculate that the finding of pleural effusions and pulmonary interstitial thickening, as seen in four of our patients, might be caused by the mediastinal masses obstructing the lymphatic channels. It would be conceivable that the underlying mechanism resembles that seen in patients with fluid overload as occurs in cardiogenic pulmonary edema. Moreover, chylothorax associated with multiple osteolytic lesions, as seen in four of our patients, was reported to carry a poor prognosis [11, 12]. Unfortunately, this prognosis was confirmed by the results of our study: all four patients displaying this combination of findings died from respiratory failure within 18 months of the initial diagnosis.
The second finding previously unreported was the extensive laryngeal involvement seen in one of our patients. Because congenital lymphangiomas often occur in the area of primitive lymph sacs [1], such as the neck or the axilla, this finding is, basically, not surprising. Unreported, however, is the extensive laryngeal infiltration with subsequent destruction of the organ, as revealed in our study by both CT and MR imaging and as confirmed by gross-pathologic workup. The fact that cervical involvement without laryngeal infiltration was seen in two of our other patients suggests that laryngeal infiltration, as seen in our patient, might correspond to a particularly advanced form of the disease.
Although rare, mesenteric thickening, as seen in four of our patients, has been sporadically described in the literature [12]. These descriptions, however, were based on surgical specimens, and no cross-sectional imaging data were provided. Imaging findings in diffuse abdominal involvement with generalized cystic alterations were reported by Cutillo et al. [2], but mesenteric thickening, as seen in four of our patients, was not described. As seen in surgical specimens, the mesenteric thickening was caused by dilated or obstructed lymphatics (or both). Evidence of splenic involvement, as seen in three of our patients both on CT and MR imaging, agrees with findings reported in the literature [2, 4, 6, 8, 10] and consisted of hypodense, confluent low-attenuation masses on CT. On MR imaging, these lesions were of variable signal intensity but appeared hypo- to isointense on T1- and hyperintense on T2-weighted images. As Cutillo et al. suggested, these heterogeneities may reflect varying ratios of chyle, water, and fat. This hypothesis was confirmed histopathologically in our patients.
Diffuse cystic involvement of bones has been extensively described [2, 4, 5, 9]. As these researchers have found, MR imaging showed these lesions better than did CT. However, CT better documented the destructive potential that finally resulted in pathologic fractures. This observation was confirmed by our results. Histopathologically, these osseous lesions represented chyle-filled and partly septated cysts.
The major drawback of our investigation is inherent to its retrospective design. The resulting heterogeneity in imaging protocols and algorithms may have contributed to some variation in the radiologic appearance of pathologic findings. The mere depiction of pathologies, however, remained untouched by our study design. Also, imaging was focused on the anatomic region where the clinical symptoms were originally located. We, therefore, might have overlooked lesions in other parts of the body. Because the retrospective design of our study essentially resulted from the rarity of generalized cystic lymphangiomatosis, we do not foresee future publication of representative prospective series. The fact that a substantial part of the literature on generalized cystic lymphangiomatosis is reports of single cases [2, 5,6,7, 9, 10] appears to confirm this assumption.
In summary, generalized cystic lymphangiomatosis appears to have a recognizable appearance on both CT and MR imaging, with genealized cystic lesions in parenchymal organs, lytic bone lesions, and diffuse thickening of pulmonary interstitium and mesenteries being the most common findings. However, our results also suggest that, due to its general distribution, generalized cystic lymphangiomatosis may be seen in virtually any organ with the exception of the nervous system, including the larynx. As seen in our patients, recognition of these findings can contribute to strengthen the diagnosis in patients with clinically suspected generalized cystic lymphangiomatosis, and, therefore, might eliminate the need for more invasive diagnostic or surgical procedures.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  H. Liu, D.R. Shatzkes, and R.A. Holliday Bilateral Orbital Lymphangiomas in Two Patients with Generalized Lymphangiomatosis AJNR Am. J. Neuroradiol., March 1, 2007; 28(3): 491 - 492. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Fox, S. K. Woods, A. Gundakaram, S. K. Varma, Y.-H. Huang, and K.-S. Wong Index of Suspicion Pediatr. Rev., July 1, 2005; 26(7): 263 - 270. [Full Text] [PDF]
|
|
|
|
 |
|
 E. M. Marom, C. A. Moran, and R. F. Munden Generalized Lymphangiomatosis Am. J. Roentgenol., April 1, 2004; 182(4): 1068 - 1068. [Full Text] [PDF]
|
|
|
|
|
|
R. Aviv, K. McHugh, P. Wunderbaldinger, and A. A. Bankier Mechanisms of Chylous Effusion in Lymphangiomatosis Am. J. Roentgenol., October 1, 2000; 175(4): 1191 - 1191. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
