TtU

Ftfoahitct

mmgpesofsba

crfb

D

A

4

he Role of 18F-FDG PET/CT inhe Evaluation of Solid Splenic Massesr Metser, MD, and Einat Even-Sapir, MD, PhD

18F-Fluorodeoxyglucose–positron emission tomography/computed tomography is a nonin-vasive imaging modality that appears to have a high negative-predictive value for malig-nancy in assessing solid splenic masses discovered on conventional imaging modalities.Histological sampling is recommended for fluorodeoxyglucose-avid splenic masses, as themain differential diagnosis includes malignancy, either primary or secondary, or granulo-matous disease.Semin Ultrasound CT MRI 27:420-425 © 2006 Elsevier Inc. All rights reserved.

rucs

paomspit(

PTPlatwep�Wtncstr

ocal solid splenic lesions, benign or malignant, are oftendiscovered incidentally. The most common benign enti-

ies are hemangioma, granuloma, and hamartoma. Focal in-arct and abscess may also occasionally appear as solid massesn imaging. Sporadically, rarer entities such as littoral cellngioma, inflammatory pseudotumor, or Gaucher nodulesave been reported.1-5 The most frequent malignancies to

nvolve the spleen are lymphoma, leukemic infiltrates, me-astases (most commonly from lung, breast, gastrointestinalancers, or melanoma), or rarely, primary angiosarcoma ofhe spleen.6-8

The morphological imaging characteristics of solid splenicasses have been described for most conventional imagingodalities. Hypovascular masses in an enlarged spleen and

eneralized lymphadenopathy may be suggestive of lym-homa. A hypervascular splenic mass, with metastatic dis-ase, may be suggestive of primary angiosarcoma of thepleen.9 When abnormalities exist in the spleen as well as inther organs, histological sampling is usually performedrom nonsplenic sites, to avoid iatrogenic injury to thepleen. However, in most cases making a definitive diagnosisased on imaging alone is difficult, especially if no otherbnormalities exist.

Splenic fine-needle biopsies are being performed with in-reasing frequency in recent years with a high success rateanging from 63 to 91%10,11 and a complication rate rangingrom 1.5 to 13%.10,12 Major hemorrhage secondary to spleniciopsy has been reported not infrequently (7.7% in one se-

epartment of Nuclear Medicine, Tel-Aviv Sourasky Medical Center, Sack-ler Faculty of Medicine, Tel-Aviv University, Tel-Aviv, Israel.

ddress reprints requests to: Ur Metser, MD, Department of Nuclear Medi-cine, Tel-Aviv Sourasky Medical Center, 6 Weizman Street, Tel-Aviv,

m64239 Israel. E-mail: umetser@tasmc.health.gov.il

20 0887-2171/06/$-see front matter © 2006 Elsevier Inc. All rights reserved.doi:10.1053/j.sult.2006.06.005

ies), requiring emergent splenectomy. Major bleeding issually a result of biopsy from vascular lesions such as littoralell angioma or angiosarcoma.10 Therefore, judicious use ofplenic needle biopsies appears prudent.

In recent years, 18F-fluorodeoxyglucose (18F-FDG)ositron emission tomography (PET) imaging has evolved aspowerful noninvasive imaging modality, mostly for oncol-gy, but also in nononcological indications, such as assess-ent of fever of unknown origin.13 In-line, hybrid PET/CT

canners have been introduced into routine clinical use in theast several years, enabling metabolic assessment of anatom-

cal abnormalities. The purpose of this review was to describehe potential role of 18F-FDG PET/computed tomographyCT) in the evaluation of solid splenic lesions.

ositron Emissionomography—Basic Principles

ET imaging is based on the use of a biochemical moleculeabeled with a positron-emitter (such as 18F, 11C, 15O) en-bling in vivo tracking of the radioactive compound. Duringhe decay process of the radioisotope, a positron is emitted,hich after traveling a short distance (about a millimeter)

ncounters an electron from surrounding molecules. The twoarticles annihilate each other, resulting in emission of two-rays of 511 keV at 180 degrees to one another (Fig 1).hen detected simultaneously (within 12 nanoseconds) by

wo opposed detectors, it is assumed that they both origi-ated from the same annihilation event located along a lineonnecting the two detectors. Data collected are recon-tructed into an image using standard algorithms. Attenua-ion correction is performed traditionally by calculating theatio between two additional scans, a blank scan and a trans-

ission scan, performed with an external positron-emitting

ratst

PThumtipcllac

1

1

acnppeodFii

plioi

y detec

Role of 18F-FDG PET/CT in the evaluation of solid splenic masses 421

otating 68Germanium rod source.14 With PET/CT scanners,ttenuating correction is performed with a fraction of theime using CT data, thus obviating the need for a transmis-ion scan, reducing acquisition time and improving patienthroughput.15

ET/CThe paucity of anatomical data from 18F-FDG-PET scansampers lesion localization and thus interpretation of focalptake of 18F-FDG. Focal uptake may be physiological oray indicate significant pathology. For example, focal up-

ake along the chest wall may be due to physiological uptaken intercostal muscles or may be due to a malignant pleurallaque. In-line hybrid PET and CT scanners have been inlinical use for several years.16 Several studies have been pub-ished on the advantages of fusion imaging, showing betteresion localization, more accurate characterization of lesionss benign or malignant, and better diagnostic accuracy asompared to side-by-side reading of PET and CT.17,18

Figure 2 Splenic infarct simulating mass lesion (CT, left;is seen in posterior aspect of spleen on CT. PET and f

Figure 1 Positron (marked “�”) encounters an electron (opposite �-rays of 511 keV. Gamma rays are detected b

consistent with an infarct. (Color version of figure is available

8F-Fluorodeoxyglucose8F-FDG is the most common radiotracer used for PET im-ging. Much like glucose, it is transported into cells by glu-ose transporter proteins and is phosphorylated by hexoki-ase to 18F-FDG-6-phosphatase. However, since 18F-FDG-6-hosphatase does not enter subsequent glucose metabolicathways and since its permeability in membranes is low, it isffectively trapped within cells. The normal biodistributionf 18F-FDG is the brain, with variable uptake in the myocar-ium, and gastrointestinal tract. The kidneys excrete 18F-DG and therefore 18F-FDG is normally seen along the gen-

tourinary tract. Physiological uptake in the liver and spleens generally uniformly low.19

Many types of tumors exhibit overexpression of trans-orter proteins, high levels of hexokinase, and decreased

evels of glucose-6-phosphatase. Therefore, these cells showncreased uptake of 18F-FDG, providing the basis for oncol-gy imaging using 18F-FDG-PET. It should however be bornen mind that 18F-FDG is not tumor-specific and increased

iddle; fused PET/CT, right). A subtle hypodense lesionET/CT images show lesion to be photopenic (arrow),

d “�”). Annihilation process results in emission of twotors (“D”) of PET scanner.

PET, mused P

marke

online.)

1

i

1

SBMhuaH1

leectt

igwc

ittktoa

odsrcw

e onlin

422 U. Metser and E. Even-Sapir

8F-FDG uptake may be seen in a variety of benign conditionsncluding inflammation.20

8F-FDG-PET/CT inolid Splenic Massesenign Solid Splenic Massesost benign solid splenic masses such as hemangiomas or

amartomas of the spleen are not expected to show abnormalptake of 18F-FDG. Focal infarcts which may on occasionppear “mass-like” are photopenic on 18F-FDG-PET (Fig 2).owever, some benign lesions exhibit abnormal uptake of

8F-FDG, most commonly splenic abscess and active granu-omatous diseases involving the spleen. Activated leukocytes,specially monocytes, show increased glucose uptake due tolevated levels of glucose transporter proteins. Furthermore,ytokines and growth factors increase the affinity of theseransporters to glucose.21,22 Granulomatous disease involvinghe spleen may be infectious (eg, tuberculosis, Brucella) or

Figure 3 Sarcoidosis. Axial PET/CT images (CT, left; PEuptake of FDG is seen in small hilar and mediastinal lymsplenic parenchyma. (Color version of figure is availabl

Figure 4 Inflammatory pseudotumor of spleen. Axial PE3.5-cm FDG-avid solid splenic mass indistinguishablfeatures and immunostains consistent with an inflamm

available online.)

nflammatory (eg, sarcoidosis). In routine clinical practice,ranulomatous disease may be suggested prospectively onhole-body imaging due to multisite involvement and typi-

al imaging patterns (Fig 3).Activated macrophages may also be the mechanism for

ncreased uptake of FDG in inflammatory pseudotumor ofhe spleen. Inflammatory pseudotumor is a benign, yet po-entially locally aggressive lesion. Although its etiology is un-nown, several factors have been implicated including infec-ion and autoimmune processes.23 It may mimic malignancyn both conventional imaging (sonography, CT, and MRI)nd 18F-FDG-PET (Fig 4).

Gaucher disease is a rare hereditary lysosomal storage dis-rder with accumulation of glucocerebroside in reticuloen-othelial cells throughout the body, resulting in hepato-plenomegaly. Splenic infarcts and focal splenic nodulesepresenting clusters of macrophage-like cells (termed “Gau-her cells”) can be found in approximately a third of patientsith type I disease on magnetic resonance (MR), likely an

dle; fused PET/CT, right). (A) Symmetrical abnormales. (B) Multiple small FDG-avid nodules are seen in thee.)

images (CT, left; PET, middle; fused PET/CT, right): amalignancy. Surgical pathology showed histologicalseudotumor of the spleen. (Color version of figure is

T, midph nod

T/CTe fromatory p

usGicpt

SMpatctbsawc

klitf6s1n

mFlamnsF

PTnomfopiimisio

edcm

lor ver

Role of 18F-FDG PET/CT in the evaluation of solid splenic masses 423

nderestimate of their true frequency on pathology.24 Notudies have been performed to date on 18F-FDG-PET andaucher nodules in the spleen. Although expression of pro-

nflammatory mediators is not always apparent on Gaucherells, markers characteristic of alternatively activated macro-hages are found.25 Therefore, Gaucher nodules may poten-ially be FDG-avid (Fig 5).

plenic Metastasesetastases in the spleen are uncommon, usually occurring as

art of disseminated metastatic disease.26 In postmortem ex-mination of cancer patients the prevalence of splenic metas-ases was between 2.3 and 12.9%, most commonly fromarcinoma of lung, breast, ovary, and melanoma.26,27 Statis-ically, most solitary splenic lesions in cancer patients areenign, as solitary splenic metastases are rare.28 Nonetheless,olitary or multiple splenic masses warrant further evaluationnd specific characterization to rule out malignancy, evenhen other metastatic disease is documented, as in some

linical settings they may alter patient management (Fig 6).A metabolically active splenic mass in a patient with

nown FDG-avid malignancy would likely represent a ma-ignant splenic deposit. Conversely, lack of 18F-FDG uptaken a splenic mass would indicate a lesion that is not related tohe patient’s known FDG-avid malignant disease and there-ore would likely be benign. In a recent study performed on8 oncology patients with FDG-avid malignancy and solidplenic masses on anatomical imaging, 18F-FDG-PET/CT had00% accuracy in characterizing lesions as benign or malig-

Figure 5 Gaucher nodule in spleen. (A) MIP PET image:enlarged spleen measuring16.5 cm in length. (B) Axialshows uptake in a 2.3-cm hypodense splenic mass. (Co

ant. In 20 additional patients without previously diagnosed 2

alignancy and a solid splenic mass, lack of uptake of 18F-DG-PET also had a 100% negative-predictive value for ma-

ignancy.29 It should however be kept in mind that non-FDG-vid tumors, such as some renal or thyroid cancers, mayetastasize to the spleen. Therefore, when encountering aon-FDG-avid splenic mass, the CT portion of the PET/CTtudy should be meticulously assessed for potential non-DG-avid malignancy.

rimary Malignancies of the Spleenhe spleen is a frequent site of involvement in patients withon-Hodgkin’s lymphoma and may be involved in one-thirdf patients with Hodgkin’s disease.24 On imaging, lympho-atous involvement of the spleen may manifest as either

ocal lesions or diffuse infiltration. A solid splenic mass mayccasionally be the presenting sign of disseminated lym-homa (Fig 7). Recent reports have shown that 18F-FDG-PET

s more accurate than CT and 67Ga-citrate scintigraphy fordentifying splenic involvement by lymphoma.30,31 Primary

alignant lymphoma of the spleen is defined as lymphomanvolving the spleen with or without lymph nodes in theplenic hilum.32 Primary malignant lymphoma of the spleens uncommon, yet it is the most common primary malignancyf the spleen.Primary nonhematopoietic tumors of the spleen may be

pithelial in origin (namely, angiosarcoma and hemangioen-othelioma) or from connective tissue (such as fibrosar-oma). Of these extremely rare entities, angiosarcoma is theost common tumor to involve the spleen with fewer than

ate focal FDG uptake (arrows) with an SUV of 4.2 in anT images (CT, left; PET, middle; fused PET/CT, right)sion of figure is available online.)

moderPET/C

00 cases reported in the literature.33 Clinically, these tumors

mmnnPab

CCMqvs

424 U. Metser and E. Even-Sapir

ay manifest with left upper quadrant abdominal pain, ane-ia, coagulopathy, and spontaneous rupture. Hematoge-ous metastases at diagnosis are common and overall prog-osis is poor.34 Although there are no reports of 18F-FDG-ET in the diagnosis of primary splenic angiosarcoma,ngiosarcoma of the pleura and chest wall has been shown toe FDG-avid.35

Figure 6 Solitary splenic metastasis in 60-year-old patienmiddle; fused PET/CT, right): an FDG-avid mass in the simiddle; fused PET/CT, right): an FDG-avid metastatic deonline.)

Figure 7 Diffuse large cell lymphoma. A 58-year-old womered incidentally on abdominal CT. Axial PET/CT imagFDG-avid solid splenic masses (arrows). (B) FDG-avid a

biopsy of groin node showed diffuse large cell lymphoma. (Co

onclusiononventional imaging modalities (sonography, CT, andRI) often discover incidental solid splenic masses. Fre-

uently, these are difficult to definitively characterize on con-entional imaging, as even hemangiomas, the most commonolid lesion in the spleen, may have a complex appearance,

igmoid cancer. (A) Axial PET/CT images (CT, left; PET,colon (arrows). (B) Axial PET/CT images (CT, left; PET,

n the spleen (arrow). (Color version of figure is available

nt for FDG-PET/CT due to solid splenic masses discov-, left; PET, middle; fused PET/CT, right). (A) Multiplenodes (not shown) and groin node (arrow). Excisional

t with sgmoidposit i

an sees (CTxillary

lor version of figure is available online.)

mocidTiea

R

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2

2

2

2

2

2

3

3

3

3

3

3

Role of 18F-FDG PET/CT in the evaluation of solid splenic masses 425

aking differentiation from malignancy difficult. Splenic bi-psy, although feasible, is associated with morbidity, espe-ially for vascular lesions. 18F-FDG-PET/CT is a noninvasivemaging modality that appears to have a high negative-pre-ictive value for malignancy in assessing solid splenic masses.he main differential diagnosis for FDG-avid splenic lesions

s malignancy (primary or secondary), or granulomatous dis-ase. Therefore, histological sampling is warranted for FDG-vid splenic masses.

eferences1. Rabushka LS, Kawashima A, Fishman EK: Imaging of the spleen: CT

with supplemental MR examination. Radiographics 14:307-332, 19942. Ramani M, Reinhold C, Semelka RC, et al: Splenic hemangiomas and

hamartomas: MR imaging characteristics of 28 lesions. Radiology 202:166-172, 1997

3. Levy AD, Abbott RM, Abbondanzo SL: Littoral cell angioma of thespleen: CT features with clinicopathologic comparison. Radiology 230:485-490, 2004

4. Variami E, Terpos E, Vgenopoulou S, et al: Inflammatory pseudotumorof the spleen: a case report and review of the literature. Ann Hematol78:560-563, 1999

5. Terk MR, Esplin J, Lee K, et al: MR imaging of patients with type 1Gaucher’s disease: relationship between bone and visceral changes. AJRAm J Roentgenol 165:599-604, 1995

6. Bragg DG, Colby TV, Ward JH: New concepts in the non-Hodgkinlymphoma: radiologic implications. Radiology 159:289-304, 1998

7. Lee SS, Morgenstern L, Phillips EH, et al: Splenectomy for splenicmetastases: a changing clinical spectrum. Am Surg 66:837-840, 2000

8. Paterson A, Frush DP, Donnelly LF, et al: A pattern-oriented approachto splenic imaging in infants and children. Radiographics 19:1465-1485, 1999

9. Thompson WM, Levy AD, Aguilera NS, et al: Angiosarcoma of thespleen: imaging characteristics in 12 patients. Radiology 235:106-115,2005

0. Keogan MT, Freed KS, Paulson EK, et al: Imaging-guided percutaneousbiopsy of focal splenic lesions: update on safety and effectiveness. AJRAm J Roentgenol 172:933-937, 1999

1. Venkataramu NK, Gupta S, Sood BP, et al: Ultrasound guided fineneedle aspiration biopsy of splenic lesions. Br J Radiol 72:953-956,1999

2. Lindgren PG, Hagberg H, Eriksson B, et al: Excision biopsy of thespleen by ultrasound guidance. Br J Radiol 58:853-857, 1985

3. Bleeker-Rovers CP, de Kleijn EM, Corstens FH, et al: Clinical value ofFDG PET in patients with fever of unknown origin and patients sus-pected of focal infection or inflammation. Eur J Nucl Med Mol Imag31:29-37, 2004

4. Maisey MM, Wahl RL, Barrington SF: Principles and Methods, in Mars-den PK (ed): Atlas of Clinical Positron Emission Tomography. London,UK, Arnold, 1999, pp 8-31

5. Messa C, Bettinardi V, Picchio M, et al: PET/CT in diagnostic oncology.

Q J Nucl Med Mol Imag 48:66-75, 2004

6. Beyer T, Townsend DW, Brun T, et al: A combined PET/CT scanner forclinical oncology. J Nucl Med 41:1369-1379, 2000

7. Antoch G, Saoudi N, Kuehl H, et al: Accuracy of whole-body dual-modality fluorine-18-2-fluoro-2-deoxy-D-glucose positron emissiontomography and computed tomography (FDG-PET/CT) for tumorstaging in solid tumors: comparison with CT and PET. J Clin Oncol22:4357-4368, 2004

8. Metser U, Golan O, Levine CD, et al: Tumor lesion detection: when isintegrated positron emission tomography/computed tomography moreaccurate than side-by-side interpretation of positron emission tomog-raphy and computed tomography? J Comput Assist Tomogr 29:554-559, 2005

9. Gallagher BM, Ansari A, Atkins H, et al: Radiopharmaceuticals XXVII:18F-labeled 2-deoxy-2-fluoro-D-glucose as a radiopharmaceutical formeasuring regional myocardial glucose metabolism in vivo—tissue dis-tribution and imaging studies in animals, J Nucl Med 18:990-996,1977

0. Love C, Tomas MB, Tronco GG, et al: FDG PET of infection and in-flammation. Radiographics 25:1357-1368, 2005

1. Zhuang H, Alavi A: 18-fluorodeoxyglucose positron emission tomo-graphic imaging in the detection and monitoring of infection and in-flammation. Semin Nucl Med 32:47-59, 2002

2. Paik JY, Lee KH, Choe YS, et al: Augmented 18F-FDG uptake in acti-vated monocytes occurs during the priming process and involves ty-rosine kinases and protein kinase C. J Nucl Med 45:124-128, 2004

3. Scott L, Blair G, Taylor G, et al: Inflammatory pseudotumors in chil-dren. J Pediatr Surg 23:755-758, 1988

4. Hill SC, Damaska BM, Ling A, et al: Gaucher disease: abdominal MRimaging findings in 46 patients. Radiology 184:561-566, 1992

5. Jmoudiak M, Futerman AH: Gaucher disease: pathological mechanismsand modern management. Br J Haematol 129:178-188, 2005

6. Berge T: Splenic metastases. Frequencies and patterns. Acta PatholMicrobiol Scand 82:499-506, 1974

7. Morgenstern L, Rosenberg J, Geller SA: Tumors of the spleen. WorldJ Surg 9:468-476, 1985

8. Agha-Mohammadi S, Calne RY: Solitary splenic metastasis: case reportand review of the literature. Am J Clin Oncol 24:306-310, 2001

9. Metser U, Miller E, Kessler A, et al: Solid splenic masses: evaluationwith 18F-FDG PET/CT. J Nucl Med 46:52-59, 2005

0. Rini JN, Leonidas JC, Tomas MB, et al: 18F-FDG PET versus CT forevaluating the spleen during initial staging of lymphoma. J Nucl Med44:1072-1074, 2003

1. Kostakoglu L, Leonard JP, Kuji I, et al: Comparison of fluorine-18fluorodeoxyglucose positron emission tomography and Ga-67 scintig-raphy in evaluation of lymphoma. Cancer 94:879-888, 2002

2. Brox A, Shustik C: Non-Hodgkin’s lymphoma of the spleen. Leuk Lym-phoma 11:165-171, 1993

3. Smith VC, Eisenberg BL, McDonald EC: Primary splenic angiosarcoma.Case report and literature review. Cancer 55:1625-1627, 1985

4. Autry JR, Weitzner S: Hemangiosarcoma of spleen with spontaneousrupture. Cancer 35:534-539, 1975

5. Del Frate C, Mortele K, Zanardi R, et al: Pseudomesotheliomatousangiosarcoma of the chest wall and pleura. J Thorac Imag 18:200-203,

2003