JOBNAME: jcat 28#6 2004 PAGE: 1 OUTPUT: Sun October 24 01:10:45 2004 lww/jcat/87535/04-277

CASE REPORT: PET IMAGING

A New Pitfall on Abdominal PET/CT: A Retained Surgical Sponge Eduard Ghersin, MD,* Zohar Keidar, MD, PhD,†‡ Olga R. Brook, MD,* Marco A. Amendola,§ and Ahuva Engel, MD*

Abstract: Positron emission tomography (PET)/computed tomography (CT) and multidetector CT findings of an abdominal retained surgical sponge (RSS) are presented. A PET/CT study performed for evaluation of a suspected abdominal tumor demonstrated the inconclusive finding of a hypometabolic area surrounded by increased 2-[fluorine-18] fluoro-2-deoxy-D-glucose uptake. Follow-up contrastenhanced multidetector CT suggested the correct diagnosis of an RSS. This is the first known report of the PET/CT appearance of an RSS. Key Words: retained surgical sponge, positron emission tomography/computed tomography, multidetector computed tomography (J Comput Assist Tomogr 2004;28:839–841)

R

etained surgical sponges (RSSs), although unusual, remain a clinical problem despite precautions taken by surgeons. Medical imaging plays an essential role in making a correct diagnosis, with a variety of cross-sectional imaging modalities used, including computed tomography (CT), ultrasonography, and magnetic resonance imaging. Although positron emission tomography (PET) or PET/CT is not used for this purpose, the differential diagnosis of some clinical presentations referred to hybrid imaging should include this entity. Therefore, one needs to be acquainted with this PET/CT pattern of RSSs.

CASE REPORT A 60-year-old man, 11 years after undergoing a splenectomy because of a myeloproliferative disease, was referred for a wholebody PET/CT scan for further metabolic characterization of a large tumor in the left upper abdomen previously diagnosed by CT. The PET/CT scan (Discovery LS; General Electric Medical Systems, Milwaukee, WI) was performed 90 minutes after the injection of 555 MBq (15 mCi) 2-[fluorine-18] fluoro-2-deoxy-D-glucose. The PET/CT images demonstrated an area of decreased uptake in the upper left abdomen localized by the CT component to the left upper abdominal mass, which was surrounded by a rim of increased 2-[fluorine-18] fluoro-2-deoxy-D-glucose activity localized by CT to

From the *Department of Diagnostic Imaging, Rambam Medical Center, Haifa, Israel, †Department of Nuclear Medicine, Rambam Medical Center, Haifa, Israel, ‡B. Rappaport Faculty of Medicine, Technion-Israel Institute of Technology, Haifa, Israel, and §Department of Radiology, University of Miami School of Medicine, Miami, FL. Reprints: Dr. Eduard Ghersin, Department of Diagnostic Radiology, Rambam Medical Center, P.O. Box 9602, Haifa 31096, Israel (e-mail: ghersine@ hotmail.com). Copyright Ó 2004 by Lippincott Williams & Wilkins

a thin hypodense capsule at the border of the mass. These findings created a smooth almost complete ‘‘ring pattern’’ on the fused images (Figs. 1B, 2A–C). The mass was initially interpreted as a possible malignant tumor of gastric origin with a large necrotic center. The patient was referred for further evaluation with contrast-enhanced multidetector CT, which was performed using an IDT 16-slice scanner (Philips Medical Systems, Cleveland, OH). The findings on the new CT scan demonstrated an enhancing peripheral capsule bordering a soft tissue density mass. In addition, an elongated stripe like a radiopaque structure located within the soft tissue mass was noted on the axial scan as well as on the scout film (see Figs. 1A, 2D). This finding was originally interpreted as a possible calcification or surgical clip on the PET/CT scan, but its elongated stripe-like shape on the axial scan and scout film enabled us to identify it as a radiopaque marker of an RSS. The diagnosis of an RSS after a splenectomy 11 years previously was made and later confirmed on explorative laparotomy.

DISCUSSION Laparotomy sponges and surgical towels are the most commonly retained foreign bodies inadvertently left after surgery since they were first reported by Wilson in 1884.1 The problem remains despite the recommended routine use of radiographically detectable sponges. Data concerning the incidence of RSSs is difficult to ascertain because of a low reporting rate; however, the estimated occurrence varies between 1 in 100 and 1 in 5000 laparotomies.2 Two different types of foreign body responses can be elicited by an RSS. The first type of foreign body reaction is exudative in nature and often leads to abscess formation. The second type leads to an aseptic foreign body granuloma with a fibroblastic reaction and incomplete encapsulation.3 This form of RSS response is usually clinically silent and may remain quiescent for a long time, although a palpable mass is sometimes found, which eventually becomes symptomatic. A spectrum of findings has been described with RSSs using a variety of imaging modalities, including plain radiography, ultrasonography, CT, and magnetic resonance imaging. The demonstration of radiopaque markers is an important clue for the diagnosis of an RSS. Nevertheless, as seen in our case on the initial CT component of the PET/CT study, radiopaque markers can be misinterpreted as calcifications, intestinal contrast material, or surgical clips. Even in an optimal situation, the plain radiograph is less than ideal, as demonstrated by Revesz et al,4 with a 10%–25% false-negative rate in experimental studies on plain films of cadavers. On CT, a typical ‘‘spongiform pattern’’ has been described in association with an RSS, reflecting bubbles of air trapped between

J Comput Assist Tomogr  Volume 28, Number 6, November/December 2004

839

JOBNAME: jcat 28#6 2004 PAGE: 2 OUTPUT: Sun October 24 01:10:47 2004 lww/jcat/87535/04-277

Ghersin et al

J Comput Assist Tomogr  Volume 28, Number 6, November/December 2004

FIGURE 1. A, Scout image from a computed tomography scan of the abdomen showing a characteristic radiopaque marker of a retained surgical sponge (RSS) in the left upper abdomen (white arrow). B, 2-[fluorine-18] fluoro-2-deoxy- D -glucose positron emission tomography coronal section of the abdomen showing an area of decreased uptake in the upper left abdomen surrounded by a thin rim of increased 2-[fluorine-18] fluoro-2-deoxy-D-glucose activity (black arrow).

840

FIGURE 2. Nonenhanced computed tomography (CT) (A), 2-[fluorine-18] fluoro-2-deoxy-D-glucose positron emission tomography (PET) (B) and fused PET/CT (C) axial images of the upper abdomen and enhanced axial CT image (D) of the upper abdomen from a separately performed CT scan. Note the soft tissue mass representing the retained surgical sponge (RSS) (*) surrounded by a thin capsule, which demonstrates mild enhancement after contrast material administration and increased 2-[fluorine-18] fluoro-2-deoxy-D-glucose uptake (solid white arrow) on PET/CT. Embedded within the RSS is a hyperdense focus representing the radiopaque marker (dotted black arrow).

q 2004 Lippincott Williams & Wilkins

JOBNAME: jcat 28#6 2004 PAGE: 3 OUTPUT: Sun October 24 01:11:03 2004 lww/jcat/87535/04-277

J Comput Assist Tomogr  Volume 28, Number 6, November/December 2004

the cotton fibers of the retained sponge.5 This pattern, although regarded as sensitive by many authors, was absent in our case. Positron emission tomography/computed tomography is a new imaging modality that is increasingly used, especially in the imaging of cancer patients. It involves a combination of the metabolic information provided by PET and the crosssectional anatomic information provided by CT, which is acquired during a single examination and fused thereafter.6,7 In our case, the RSS induced an aseptic foreign body granuloma with a fibroblastic reaction and incomplete encapsulation. The RSS manifested itself on the PET component as an area of decreased uptake with a radiopaque marker incorporated within that area on the fused image. The thin capsule surrounding the RSS demonstrated increased 2-[fluorine-18] fluoro-2-deoxy-D-glucose activity, indicating the presence of metabolic activity related to its fibroblastic content. A tumor with prominent central necrosis or hemorrhage could have shown a similar presentation, although it would not have an incorporated radiopaque marker.8,9 This case report is the first description of the imaging pattern of RSSs with aseptic foreign granulomas on PET/CT. The demonstration of increased 2-[fluorine-18] fluoro-2-deoxy-D-glucose uptake surrounding an inner mass of soft tissue density void of 2-[fluorine-18] fluoro-2-deoxy-D-glucose uptake should alert nuclear medi-

q 2004 Lippincott Williams & Wilkins

PET/CT Appearance of Retained Surgical Sponge

cine specialists and radiologists to the possibility of an RSS. The finding of a radiopaque marker within the inner soft tissue mass is an additional important clue that should help in avoiding this diagnostic pitfall on PET/CT. REFERENCES 1. Williams RG, Bragg DG, Nelson JA. Gossypiboma—the problem of the retained surgical sponge. Radiology. 1978;129:323–326. 2. Wolfson KA, Seeger LL, Kadell BM, et al. Imaging of surgical paraphernalia: what belongs in the patient and what does not. Radiographics. 2000;20:1665–1673. 3. Kopka L, Fischer U, Gross AJ, et al. CT of retained surgical sponges (textilomas): pitfalls in detection and evaluation. J Comput Assist Tomogr. 1996;20:919–923. 4. Revesz G, Siddiqi TS, Buchheit WA, et al. Detection of retained surgical sponges. Radiology. 1983;149:411–413. 5. Parienty RA, Pradel J, Lepreux JF, et al. Computed tomography of sponges retained after laparotomy. J Comput Assist Tomogr. 1981;5:187–189. 6. Cook GJ, Wegner EA, Fogelman I. Pitfalls and artifacts in 18FDG PET and PET/CT oncologic imaging. Semin Nucl Med. 2004;34:122–133. 7. Kapoor V, McCook BM, Torok FS. An introduction to PET-CT imaging. Radiographics. 2004;24:523–543. 8. Cook GJ, Fogelman I, Maisey MN. Normal physiological and benign pathological variants of 18-fluoro-2-deoxyglucose positron-emission tomography scanning: potential for error in interpretation. Semin Nucl Med. 1996; 26:308–314. 9. Bhargava P, Zhuang H, Kumar R, et al. Ring-shaped FDG uptake in the right lower lung: is it always a tumor? Clin Nucl Med. 2004;29:324–325.

841