| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Gastrointestinal Imaging |
1 From the Departments of Radiologic Pathology (A.D.L.) and Hematopathology Pathology (S.L.A.), Armed Forces Institute of Pathology, 6825 16th St NW, Washington, DC 20306-6000; Department of Radiology and Nuclear Medicine, Uniformed Services University of the Health Sciences, Bethesda, Md (A.D.L., R.M.A.); and Department of Radiology, University of Maryland School of Medicine, Baltimore (R.M.A.). Received February 5, 2003; revision requested April 23; revision received May 1; accepted June 23. Address correspondence to A.D.L. (e-mail: levya@afip.osd.mil).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Two abdominal radiologists retrospectively reviewed the contrast material–enhanced CT images obtained in six, the contrast-enhanced and nonenhanced CT images obtained in two, and the photographs of gross pathologic specimens resected from seven patients. They also retrospectively reviewed clinical data (ie, demographic data, presenting signs and symptoms, physical findings, and medical histories). Histopathologic specimens from the eight patients were reviewed by a hematopathologist. The CT images were reviewed for the presence of splenomegaly. The number, size, and enhancement characteristics of the splenic masses at CT were compared with the histopathologic and gross pathologic specimen findings.
RESULTS: All patients had laboratory evidence of hypersplenism. Seven patients (88%) had splenomegaly and innumerable splenic masses ranging from 0.2 to 6.0 cm in diameter at CT. The single patient with a normal spleen size had four splenic masses. The splenic masses were hypoattenuating relative to the normal spleen at CT in all patients and correlated with blood-filled nodules at gross pathologic examination and with blood-filled vascular channels of littoral cell angioma at histopathologic examination. The early and late portal venous phase CT images that were available in one case demonstrated progressive homogeneous contrast enhancement of the masses such that they were indistinguishable from the normal splenic parenchyma.
CONCLUSION: Littoral cell angioma is a primary splenic neoplasm that most commonly manifests at CT as multiple hypoattenuating masses in an enlarged spleen. Histopathologically, these masses represent blood-filled vascular channels.
Index terms: Computed tomography (CT), contrast enhancement, 775.12112, 775.12114, 775.12115 • Spleen, CT, 775.12112, 775.12114, 775.12115 • Spleen, neoplasms, 775.3194
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
In most published cases, patients with littoral cell angioma present with splenomegaly and signs and symptoms of hypersplenism. Littoral cell angioma was originally thought to be a benign lesion. However, one patient in a reported series had disease disseminated to the liver and brain that responded to cytotoxic chemotherapy (1). Two subsequent case reports describe variants of littoral cell angioma that have histologic features of malignancy (2,3). Therefore, the malignant potential of littoral cell angioma has not been firmly established in the literature.
To our knowledge, there have been 11 cases of littoral cell angioma reported in the English-language literature since the description of this disease in 1991. These reports include isolated case reports describing the ultrasonographic, computed tomographic (CT), and magnetic resonance imaging features of littoral cell angioma published in the more recent radiology literature (4–9). The purpose of our study was to evaluate the clinical, pathologic, and CT features of littoral cell angioma of the spleen in a series of eight patients.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
The study population consisted of five women (age range, 36–84 years; mean age, 57.6 years) and three men (age range, 32–61 years; mean age, 44.7 years). The entire population ranged in age from 32 to 84 years (mean age, 52.8 years; median age, 48 years). Seven patients were white, and the remaining patient was of unknown ethnicity.
Clinical Data and Histopathologic Review
Clinical and laboratory data were reviewed by two experienced abdominal radiologists (A.D.L., R.M.A.). Clinical data were reviewed for the patients’ age, sex, race, presenting signs and symptoms, clinical evidence of splenomegaly, and medical history of malignancy. Laboratory data were reviewed for platelet count and hemoglobin and hematocrit levels as evidence of hypersplenism. All patients underwent splenectomy. Histopathologic records and specimens were available for all patients. All histopathologic material was re-reviewed by an experienced hematopathologist (S.L.A.). The hematoxylin-eosin–stained slide material was reviewed in each case to reconfirm the morphologic diagnosis of littoral cell angioma. The morphologic diagnosis was based on the presence of anastomosing vascular channels lined with tall endothelial cells, focal papillary fronds, and normal splenic sinuses at the periphery of the lesion (1). The slides were also assessed for the presence of hemosiderin, calcification, extramedullary hematopoiesis, and capsule formation. No immunohistochemical data were available for review in any of the cases. The pathologic records of each patient were reviewed by all of the authors to establish the splenic sizes and immunohistochemical values reported herein.
CT Image Acquisition and Review
Abdominal CT images obtained in all patients were available: Only contrast material–enhanced CT images were available for six and both contrast-enhanced and nonenhanced CT images were available for two patients. By using the hepatic enhancement pattern as a guide, it was determined that three patients had been scanned during only the early portal venous phase of contrast enhancement, one had been scanned during both the early portal venous phase and the late portal venous phase, and four had been scanned during the late portal venous phase only. Early portal venous phase was defined as the phase when enhancement of the portal veins was greater than that of the adjacent nonenhancing liver parenchyma. Late portal venous phase was defined as the phase during mild enhancement of the liver with respect to the portal vein.
All CT scanning had been performed by using helical scanners and section thicknesses ranging from 5 to 10 mm. Because the patients were referred from many institutions, their CT examinations had been performed by using a variety of equipment and differing section thickness and contrast material injection protocols. Thus, the CT examinations were not standardized.
Two abdominal radiologists (A.D.L., R.M.A.), each with 10 years of experience, reviewed all of the images retrospectively, with final interpretations made by consensus. Digital CT images obtained in each patient were reviewed on a computer monitor (ViewSonic, Walnut, Calif). Images displayed in soft-tissue window settings were available. CT workstation functionality to adjust window and level settings was not available on the computer monitor.
Only the histopathologic diagnosis for each patient was known at the time of CT image interpretation. The interpreting radiologists were blinded to the clinical data and gross pathology results at the time of initial CT image interpretation. The CT images were reviewed for the presence of splenomegaly, which was deemed to be present when the maximal splenic length exceeded 12 cm or the overall splenic dimensions exceeded 12 x 7 x 4 cm (10). The splenic volume was calculated by using the splenic volumetric index, which is the product of splenic length times splenic width times splenic anteroposterior thickness, all divided by 27 (Table) (10). The normal splenic volumetric index ranges from 8 to 34 cm2.
|
After the CT images were reviewed, the two interpreting radiologists (A.D.L., R.M.A.) compared them with the clinical data, pathology reports, photographs of resected gross specimens (from six patients), and intraoperative photographs (for one patient). The appearance (ie, size, shape, margin, capsule formation) of the lesion on the gross specimen photograph was compared with its appearance on the CT image. The photographs were also evaluated for evidence of hemorrhage within the lesion.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
The single patient who did not have clinical evidence of splenomegaly was evaluated for persistent rectal bleeding and anemia 2 weeks after undergoing colonoscopy. All patients underwent splenectomy. Only one patient had preexisting hematologic disorders. This patient had chronic idiopathic thrombocytopenia purpura and myelodysplastic syndrome. One patient had a history of malignancy, which was a bronchial adenocarcinoma.
Histopathologic and Immunophenotypic Features
All patients met the criteria for the diagnosis of littoral cell angioma established by Falk et al (1). Multiple lesions were confirmed histopathologically in all patients. The lesions were located in the splenic red pulp and composed of variably sized vascular channels lined with flat and tall endothelial cells (Fig 1). Papillary fronds extended into the vascular channels (Fig 1b). Marked hemosiderin pigmentation was identified in two patients. Calcification was histopathologically identified in one patient, and associated extramedullary hematopoiesis was identified histopathologically in another patient. Neither the calcification nor the extramedullary hematopoiesis was evident at CT because the findings were microscopic at histopathologic examination. There was no evidence of capsule formation in any of the lesions; however, all of the lesions were well demarcated from the normal splenic parenchyma. The immunohistochemical finding (from review of pathology records) of reactivity with factor VIII, lysozyme, and CD68 antigens in six patients supported the diagnosis of littoral cell angioma (1).
|
|
|
|
|
In the two patients for whom unenhanced CT images were available, the splenic masses were not visible on the unenhanced images, but they became hypoattenuating with respect to the normal attenuating splenic parenchyma during the late portal venous phase of enhancement. In all four patients in whom early portal venous phase CT images were obtained, the images showed the lesions to have homogeneous low attenuation with respect to the normal spleen. However, the lesions in two of these patients also showed heterogeneous high attenuation (Fig 4). The lesions were hypoattenuating on the late portal venous phase images obtained in all except one patient. In this patient, the early portal venous phase images showed hypoattenuating lesions that became isoattenuating with the adjacent spleen tissue during the late portal venous phase (Fig 5).
|
|
|
|
Gross Pathologic Features
All available photographs of gross pathologic specimens (for seven patients) showed focal nodules on the cut surface of the spleen that correlated with the CT features. The nodules were dark red, brown, or black (Table)—pigments consistent with blood or blood products of varying chronicity (Fig 3b). Although there was no evidence of a capsule, the nodules were well delineated from the adjacent normal splenic parenchyma in each case. In three patients, there were multiple tiny (ie, <1–2 mm in diameter) cystic spaces associated with each nodule; histopathologically, these spaces represented the blood-filled channels of littoral cell angiomas. These small cystic spaces could not be identified at CT.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
The consistent CT feature was the presence of splenic masses too numerous to count, which were present in all but one patient. The CT attenuation pattern observed in the series was similar to that previously reported: low-attenuating lesions on contrast-enhanced images (4,5,7,8,11).
There was an inherent limitation of our study based on the nature of the patients in the referral population. The examinations were performed with different equipment and protocols at various institutions over an 8-year period. Although all of the examinations were performed with modern helical CT scanners, the lack of standardization of the contrast material injection techniques could not be overcome in a study population of this type. This is of utmost relevance because the focus of the study was the evaluation of a vascular neoplasm in a highly vascular organ. The findings in three cases in our series indicated that imaging the spleen during multiple phases of contrast enhancement may aid in the diagnosis of littoral cell angioma.
We did not have the advantages of being able to review images with narrow window settings or use workstation functions to alter windows and levels. Subtle splenic lesions can be overlooked with routine soft-tissue window settings. However, we had the advantage of being able to review photographs of resected gross specimens and pathologic records to ensure that our CT interpretations were correct.
In the two patients for whom both unenhanced and contrast-enhanced CT images were available, the lesions were not apparent until contrast material was administered. During the early portal venous phase of contrast enhancement, littoral cell angioma lesions have low attenuation with respect to the normally enhancing splenic parenchyma. Furthermore, the images obtained in the one patient who underwent both early and late portal venous CT showed the littoral cell angioma lesions had delayed contrast material filling such that they became isoattenuating with the adjacent spleen tissue. Kinoshita et al (4) and Ring and Corse (8) had published cases with the same enhancement pattern. This finding suggests that the vascular channels that form littoral cell angioma become perfused with contrast material at a slower rate than does the normal splenic parenchyma.
The histopathologic definition of littoral cell angioma is based on the presence of anastomosing vascular channels lined with tall endothelial cells and papillary fronds. In three cases in this study, the vascular channels were identified as small (ie, <1–2 mm) cystic spaces in the gross pathologic specimen. These spaces could not be identified on the corresponding CT images, most likely because of the small size of the cystic spaces as compared with the CT section thickness.
The normal spleen enhances in a mottled or striped pattern during the arterial phase of contrast enhancement (ie, peak aortic enhancement) owing to variable rates of flow through the cords of the red pulp (12,13). In some instances, the normal striped pattern of early arterial enhancement makes it difficult to interpret the presence of a splenic mass. In our study, the masses were easily identified in the four patients for whom portal venous phase images were available. We believe that this is because the spleens in this patient series were markedly enlarged and the masses were round or irregularly shaped. In contrast, a normal spleen has a linear or striped arterial enhancement pattern that becomes more homogeneous in the late portal venous phase.
The differential diagnosis for multiple low-attenuating splenic lesions at CT includes many neoplastic and nonneoplastic disorders. Lymphoma, metastatic disease, multiple hemangiomas, and, rarely, lymphangiomas may have a CT appearance that is indistinguishable from that of littoral cell angioma. Hemangiomas are the most common benign primary neoplasms of the spleen (14). At CT, they are typically hypoattenuating masses that may have a contrast enhancement pattern similar to that of hepatic hemangiomas or a pattern of mottled contrast enhancement on delayed images (15). Occasionally, calcification, cysts, and/or fibrosis may be present within the hemangioma. Diffuse involvement of the spleen, or hemangiomatosis, is rare.
Lymphoma and metastatic disease frequently affect the spleen in a diffuse manner that is indistinguishable from the pattern observed in littoral cell angioma. Additional findings such as adenopathy or disease in other organs may aid in the diagnosis of lymphoma or metastatic disease (16,17).
The nonneoplastic disorders in the differential diagnosis of littoral cell angioma include infections such as disseminated fungal disease, septic emboli, and granulomatous diseases such as sarcoidosis and tuberculosis. Findings of associated adenopathic, pulmonary, and mediastinal diseases should aid in the diagnosis of sarcoidosis (18). In immunocompromised individuals, tuberculosis, Mycobacterium avium-intracellulare complex, Pneumocystis carinii, and disseminated Kaposi sarcoma may cause multiple splenic masses (19,20). In these cases, the patient’s clinical condition and immune status will help distinguish an infectious entity from a neoplastic disorder such as littoral cell angioma.
The diagnosis of littoral cell angioma with fine-needle aspiration biopsy has rarely been reported in the medical literature (21). Preoperative knowledge of the diagnosis of littoral cell angioma may be useful in the treatment of those patients who do not have clinically important hypersplenism. Imaging-guided percutaneous biopsy of the spleen is often avoided because the spleen is highly vascular. However, in relatively recently published reviews on imaging-guided percutaneous biopsy of the spleen, successful diagnoses for 91.0% (22) and 88.9% (23) of patients are reported.
The reported complication rate with imaging-guided percutaneous biopsy is 0%– 2% in most series (23). A higher complication rate (10.3%) has been reported for patients with refractory thrombocytopenia or vascular splenic neoplasms (22). Because patients with littoral cell angioma frequently have thrombocytopenia, a careful assessment of their clotting function should be made prior to attempts at percutaneous biopsy.
In conclusion, littoral cell angioma is a rare splenic neoplasm composed of multiple blood-filled vascular channels. The most common CT manifestation of littoral cell angioma is splenomegaly with innumerable masses. Littoral cell angioma should be considered in the differential diagnosis of multiple splenic masses, particularly if the patient has clinical symptoms of hypersplenism. The presence of homogeneous enhancement of the lesions on delayed contrast-enhanced CT images may aid in the differential diagnosis.
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, A.D.L., R.M.A., S.L.A.; study concepts and design, A.D.L., R.M.A., S.L.A.; literature research, A.D.L.; clinical studies, A.D.L., R.M.A.; data acquisition and analysis/interpretation, A.D.L., R.M.A., S.L.A.; manuscript preparation, A.D.L.; manuscript definition of intellectual content, editing, revision/review, and final version approval, A.D.L., R.M.A., S.L.A.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Bhatt, J. Huang, and V. Dogra Littoral Cell Angioma of the Spleen Am. J. Roentgenol., May 1, 2007; 188(5): 1365 - 1366. [Full Text] [PDF]
|
|
|
|
 |
|
 S. Bhatt, R. Simon, and V. S. Dogra Littoral Cell Angioma: Sonographic and Color Doppler Features J. Ultrasound Med., April 1, 2007; 26(4): 539 - 542. [Full Text] [PDF]
|
|
|
|
 |
|
 Image of the Month--Diagnosis Arch Surg, November 1, 2005; 140(11): 1128 - 1128. [Full Text] [PDF]
|
|
|
|
 |
|
 J. D. Floyd, P. A. Kaplan, E. R. Sauter, M. C. Perry, and D. C. Doll Patients With Unusual Bladder Malignancies and a Rare Cause of Splenomegaly: CASE 3. Littoral Cell Angioma of the Spleen in a Patient With Previous Lymphoma J. Clin. Oncol., July 1, 2005; 23(19): 4460 - 4462. [Full Text] [PDF]
|
|
|
|
 |
|
 W. M. Thompson, A. D. Levy, N. S. Aguilera, L. Gorospe, and R. M. Abbott Angiosarcoma of the Spleen: Imaging Characteristics in 12 Patients Radiology, April 1, 2005; 235(1): 106 - 115. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. Metser, E. Miller, A. Kessler, H. Lerman, G. Lievshitz, R. Oren, and E. Even-Sapir Solid Splenic Masses: Evaluation with 18F-FDG PET/CT J. Nucl. Med., January 1, 2005; 46(1): 52 - 59. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. M. Abbott, A. D. Levy, N. S. Aguilera, L. Gorospe, and W. M. Thompson From the Archives of the AFIP: Primary Vascular Neoplasms of the Spleen: Radiologic-Pathologic Correlation RadioGraphics, July 1, 2004; 24(4): 1137 - 1163. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
