| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Clinical Observations |
1 Department of Radiology, Universitair Ziekenhuis Antwerpen, University of
Antwerp, B-2650 Edegem, Belgium.
2 Present address: Department of Radiology, General Hospital Sint-Lucas,
Groenebriel 1, Ghent B-9000, Oost-Vlaanderen, Belgium.
3 Department of Radiology, Division of Abdominal Imaging and Intervention,
Brigham and Women's Hospital, Harvard Medical School, Boston, MA 02115.
4 Department of Internal Medicine, Ziekenhuisnetwerk Antwerpen, B-2060 Antwerp,
Belgium.
5 Present address: Intensive Care Unit, Royal Darwin Hospital, Rocklands, Tiwi,
Northern Territory, Australia.
Received January 7, 2005;
accepted after revision February 22, 2005.
Address correspondence to A. I. De Backer
(adelard.debacker{at}azstlucas.be).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CONCLUSION. Tuberculous focal splenic lesions are small, multiple, and most often associated with splenomegaly. Signal intensities vary on both T1- and T2-weighted images. Two different enhancement patterns are noted: peripheral enhancement and gradual peripheral enhancment with complete fill-in.
Keywords: abdominal imaging • MRI • spleen • splenomegaly • tuberculosis
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Diagnosis of tuberculosis was made by a highly specific polymerase chain reaction-based (PCR) assay for mycobacterial DNA on bronchoalveolar lavage fluid (n = 3), fine-needle aspiration biopsy of focal liver lesion (n = 2), and surgically obtained biopsy specimen of a peripheral (n = 2) and an abdominal (n = 1) lymph node. In all eight patients, the diagnosis of tuberculosis was confirmed by growth of M. tuberculosis on culture.
The presenting symptoms of the patients with tuberculosis consisted of loss of body weight (n = 7), fever (n = 6), abdominal pain (n = 4), cough (n = 4), headache (n = 4), vomiting (n = 3), general weakness (n = 3), night sweats (n = 2), anorexia (n = 1), dyspnea (n = 1), and painful mass in the forefoot (n = 1). The duration of symptoms ranged from 1 day to several months. In one patient with AIDS disease, esophageal candidiasis was a manifestation of a concurrent opportunistic infection.
In all patients, extrasplenic abdominal involvement by tuberculosis was noted: lymph nodes (n = 7), liver (n = 5), urogenital tract (n = 3), peritoneum (n = 4), gallbladder (n = 1), pancreas (n = 2), and small bowel (n = 1). Other sites of involvement consisted of peripheral lymph nodes (n = 2), spine (Pott's disease; n = 1), abdominal wall (n = 1), thoracic wall (n = 1), CNS (n = 1), and osteoarticular system (n = 1). Active pulmonary tuberculosis was present in seven patients.
MRI Technique
MRI studies were performed with a 1.5-T magnet (Signa EchoSpeed Plus, GE
Healthcare) using a torso phased-array coil. Thin-section T2-weighted
single-shot fast spin-echo images (TR/TE, infinite/60; section thickness, 4
mm) without and with fat suppression were obtained. T1-weighted
gradient-recalled echo images without and with fat suppression were obtained
with a TR of 105 msec, a TE of 1.6 msec, 256 x 512 matrix, a rectangular
field of view to reduce the number of phase-encoding views, 5-mm section
thickness with a section gap of 2 mm or less, and a flip angle of 60°.
Contrast-enhanced gradient-recalled echo images were obtained 25 sec (arterial
phase), 60 sec (portal venous phase), and 3 min (equilibrium phase) after IV
administration of 0.1 mmol per kilogram of body weight of gadobenate
dimeglumine (Multihance, Bracco-Altana) at 2 mL/sec, followed by a 20-mL
saline flush at the same rate. Gadobenate dimeglumine, a gadolinium chelate
with a hepatobiliary phase, was not used in a way different from conventional
extracellular fluid gadolinium contrast agents and, therefore, resulted in the
same imaging behavior.
Image Analysis
MR images of the spleen were reviewed by two radiologists in consensus with
regard to splenic volume (cm3) and the number of tuberculous focal
lesions, their morphology (nodular, sharply delineated, ill defined), size
(mm), their signal intensities in relation to adjacent splenic tissue, and
their enhancement pattern. A focal lesion was defined as an area with abnormal
signal intensity, contrast enhancement pattern, or both in relation to normal
surrounding splenic tissue. A lesion size of 10 mm or less was classified as
micronodular; when larger, as macronodular.
Splenic volume (cm3) was calculated according to the following
formula:
 |
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
|
|
|
|
|
|
On T2-weighted images, iso-, hyper-, and hypointense lesions were noted in three (37.5%), two (25%), and two (25%) patients, respectively. In one patient (12.5%), both hyper- and hypointense lesions were noted. On T1-weighted images, lesions were hypo-, iso-, and hyperintense in four patients (50%), three patients (37.5%), and one patient (12.5%), respectively. On dynamic contrast-enhanced images, two enhancement patterns were noted: peripheral enhancement (six patients) and gradual peripheral enhancement with complete fill-in (one patient). In the remaining patient, lesions with both enhancement patterns were noted. Splenic lesions larger than 5 mm in diameter showed the peripheral enhancement pattern only. Both peripheral enhancement and gradual peripheral enhancement with complete fill-in patterns were shown with smaller lesions.
After initiation of antituberculous chemotherapy, focal splenic lesions and other manifestations of tuberculosis showed marked or complete regression on follow-up CT or MRI studies (or both) in all patients.
Correlation of MRI Findings with CT and Sonography Findings
All eight patients underwent CT or sonography of the abdomen 1-8 days
(mean, 3.5 days) before undergoing the MRI investigation.
Five patients underwent abdominal helical CT (ProSpeed SX Advantage, GE Healthcare; slice thickness, 7 mm). Contrast-enhanced images were obtained in all patients 70 sec after the start of an IV injection of 120 mL of contrast material (meglumine ioxithalamate, 300 mg I/mL [Telebrix 30, Guerbet]) delivered at a rate of 2.5 mL/sec using a mechanical power injector. In all patients, tuberculous focal splenic lesions were visible as sharply delineated, nodular, hypodense masses showing peripheral enhancement. The size of the visible lesions on CT was identical to that on MRI (range, 3-10 mm; mean, 5 mm). In two patients, fewer splenic lesions were visible on CT (range, 4-22 lesions; mean, 11.2 lesions) compared with MRI (range, 4-36 lesions; mean, 14.4 lesions). On MRI, the latter lesions were small (3 mm) and in one patient showed complete fill-in after gadolinium administration.
Three patients underwent abdominal sonography. Examinations were performed by different examiners with a high-resolution sonography system (Sonoline Elegra, Siemens Medical Solutions) using a 3- to 5-MHz convex array broadband transducer. Splenomegaly was noted in two patients, and normal splenic volume was described in one patient. In all three patients, no focal splenic lesions were delineated.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In all our patients, diagnosis of splenic tuberculosis was made on the basis of a PCR-based assay for mycobacterial DNA, isolating M. tuberculosis from organs other than the spleen, and the regression of these lesions after antituberculous chemotherapy. Because of the absence of revealing material for histopathologic examination from a splenic lesion, we could not entirely exclude the possibility that in some cases the splenic lesions represented a synchronous distinct process. However, based on previous reports, fine-needle aspiration cytology of the spleen is not required for a final diagnosis of splenic tuberculosis when M. tuberculosis has been isolated from other organs and follow-up imaging shows a reduction in the number and size of lesions, development of calcifications within the lesions, or complete regressions of the lesions [4].
On surgical pathology, findings of splenic involvement with disseminated tuberculosis have been reported to correspond to splenic hypertrophy and the presence of miliary granulomas throughout the splenic parenchyma. In patients with disseminated tuberculosis, the weight of the spleen may reach up to 1,350 g. On section, numerous tubercles may be seen or they may be so small that they are not visible to the naked eye. When larger nodules are present, a mottled appearance may occur, with caseous yellow areas interspersed throughout the congested pulp [9].
Sonography and CT are valuable imaging methods in the evaluation of splenomegaly and the detection of focal splenic lesions. However, MRI plays an increasing role in the evaluation of patients with pathologic conditions of the abdomen.
On imaging, two main types of splenic tuberculosis have been described: the micronodular and macronodular forms [2, 4, 5, 10]. The more common micronodular form usually manifests as only moderate splenomegaly. The individual lesions typically are below the resolution capability of imaging techniques, but may appear on CT as tiny low-density foci throughout the spleen.
The macronodular form is extremely rare and may be seen as solitary or multiple rounded or oval splenic lesions measuring between 1 and 3 cm [5]. The lesions rarely exceed more than 3 cm in diameter. However, a few giant lesions have also been reported with a diameter of up to 17 cm [11]. In the reported cases, the CT findings of the macronodular form are considered variable and nonspecific. Findings vary from a noncalcified low-density lesion to a calcified high-density lesion with or without rim enhancement. Calcifications have been reported to occur in later stages of the disease [12, 13].
On MRI, a solitary tuberculoma may also show nonspecific features. A tuberculoma has been described on T1-weighted images as an isointense lesion compared with adjacent splenic tissue and may become visible on T2-weighted images as a hypointense mass with hyperintense areas. On gadolinium-enhanced images, slight peripheral rim enhancement may be seen [5].
On MRI, tuberculous focal splenic lesions may show variable signal intensities and enhancement patterns after IV administration of gadolinium. This spectrum of variable imaging findings may represent different phases of disease progression corresponding to different degrees of fibrosis, granuloma formation, caseation, and liquefaction necrosis [5]. Lesion hypointensity on T2-weighted images is thought to be due to the presence of free radicals produced by macrophages during active phagocytosis, may be associated with increased fibrosis and granulomatous tissue, or may reflect the presence of calcifications [2, 14]. The finding of lesion hypointensity on T2-weighted images may be a helpful characteristic for differentiating splenic tuberculoma from other neoplastic or inflammatory lesions [6]. Furthermore, a hypointense nodule with a less hypointense rim on T1-weighted images and a hyperintense central area with a less intense rim on T2-weighted images have also been reported [6]. These findings reflect histologic differences between the center and the periphery of the tuberculous lesion. Based on previous reports, these findings may reflect caseating granuloma with a liquid center and peripheral reactive fibrosis [15]. Finally, a hyperintense mass without rim hypointensity on T2-weighted images may also be noted. The latter finding may reflect extensive central liquefaction necrosis with only minimal peripheral granuloma formation or fibrosis (or both).
In our study, a characteristic finding on dynamic contrast-enhanced MRI consisted of a central unenhanced lesion with peripheral enhancement. This finding probably represents central caseation or liquefaction necrosis with peripheral granulation tissue. A less common pattern consisted of a peripheral enhancing lesion with, on delayed images, complete fill-in. The latter finding probably represents a granuloma with minimal or absent caseation necrosis.
The differential diagnosis of multiple focal splenic lesions includes lymphoma, Kaposi's sarcoma, metastases, sarcoidosis, bacillary angiomatosis, fungal pyogenic abscesses, histoplasmosis, and disseminated Mycobacterium avium-intracellulare and Pneumocystis carinii infections. Warshauer et al. [16] distinguished three major categories of patients presenting with multiple focal splenic lesions on the basis of the differences in symptomatology and the notion of known primary malignancy. In the symptomatic patient without known malignancy, one should mainly think of lymphoma, infection, and sarcoid. In the asymptomatic patient without known malignancy, benign tumor, sarcoid, and nonlymphomatous metastatic disease should be considered, whereas in the patient with known malignancy multiple focal lesions may represent metastases. Good differential diagnostic parameters are lesion size and low signal intensity of the lesions on T2-weighted MRI. Although overlap exists among the diagnostic groups, lymphoma tends to have larger, more variable nodules, whereas infection tends to occur with smaller, more uniform nodules. Sarcoid is intermediate in appearance. The low signal intensity on the T2-weighted images may help in the differential diagnosis with infection. Moreover, infection appears usually in the context of suppression of the immune system. Bacillary angiomatosis of the spleen is another rare condition of the spleen that may appear indistinguishable from mycobacterial infection [17]. Splenic P. carinii infection and Kaposi's sarcoma should be considered when focal lesions are detected in an immunodeficient patient. P. carinii lesions have a characteristic tendency to calcify in a rim-like or punctate fashion [17].
A patient's history and presenting symptoms; laboratory tests including blood cultures; the presence of abnormalities in other organs; and the distribution, size, signal intensities, and patterns of contrast enhancement of nodes may be helpful in the differential diagnosis. When noninvasive methods fail to provide a definite diagnosis, fine-needle aspiration biopsy of an easily accessible lesion—for example, lymph node or liver—should be considered. Diffuse splenic enlargement without focal lesions was previously thought to be a contraindication to invasive procedures. However, fine-needle aspiration biopsy of the spleen has recently been reported to be of diagnostic value with a low complication rate [18].
In conclusion, splenic involvement in our population of patients with disseminated tuberculosis consists of a combination of nodular focal lesions and splenomegaly. Signal intensities vary both on T1- and T2-weighted images and probably reflect different stages of the tuberculous process. On contrast-enhanced T1-weighted images, lesions most often show peripheral enhancement. Less commonly, gradual peripheral enhancement with complete fill-in may be seen.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
