| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
by the American Institute of Ultrasound in Medicine J Ultrasound Med 22:413-417 • 0278-4297
Contrast-Enhanced Ultrasonographic Detection of Small Pancreatic InsulinomaDepartments of Radiology (M.D., G.M., S.V., C.P.) and Surgery (M.F.), University of Verona, Verona, Italy. Address correspondence and reprint requests to Mirko D’Onofrio, MD, Department of Radiology, University Hospital G. B. Rossi, Piazza L. A. Scuro 10, 37134 Verona, Italy.
Abbreviations: CT, computed tomography • MR, magnetic resonance
Endocrine tumors of the pancreas originate from multipotential stem cells that have retained the capacity to proliferate and differentiate themselves in the various cellular lines that make up this group of neoplasms. Insulinomas represent the most frequently found functioning endocrine tumors of the pancreas and are benign in most cases (85%–99%), are single (93%–98%), and have diameters of less than 2.5 cm.1,2 When malignant, their diameters are generally greater than 3 cm, and about one third of these have metastases at the moment of diagnosis.1,2 They appear early with signs and symptoms linked to hypoglycemia due to the hypersecretion of insulin; they are almost exclusively located at the pancreatic level with no preferential sites. Preoperative ultrasonographic detection of insulinomas is generally difficult but possible in 25% to 60% of cases.3 The main difficulty in detecting a pancreatic insulinoma lies in the organ’s anatomic structure and the tumor’s small dimensions at clinical appearance. The use of second-generation contrast-enhanced imaging can increase the capacity to detect pancreatic insulinomas in preoperative ultrasonography, as the following case confirms.
A 40-year-old woman was brought to our attention because of typical symptoms of hypoglycemic crisis episodes recurring for about 1 year, 1 of which had led to a hypoglycemic coma. Blood analysis revealed a serum glucose level of 2.2 mmol/L (40 mg/dL) and a corresponding serum insulin level of 103.3 pmol/L (14.4 µU/ mL), data that supported the diagnostic indication of an insulinoma. We were asked to locate it. The ultrasound examination showed no expansive formations in the pancreatic area (Fig. 1A  ). A subsequent computed tomography (CT) investigation in the arterial phase did not reveal any images resembling hypervascularized expansive formations (Fig. 1B), whereas in another hospital, a magnetic resonance (MR) imaging study was not conclusive.
A contrast-enhanced ultrasonographic study of the pancreas was performed. Contrast medium was used to study the pancreatic gland to detect any isoechoic expansive formation with a vascularization different from the rest of the parenchyma. Sulfur hexafluoride was used in the form of microbubbles (SonoVue; Bracco SpA, Milan, Italy), and harmonic microbubble-specific imaging with low ultrasonic acoustic intensity (2- to 4-MHz coherent contrast imaging; mechanical index, <0.2; frame rate, 12–13 frames per second) was performed on an Acuson Sequoia 512 system, version 6.0 (Siemens Medical Solutions, Mountain View, CA).
For studying all the pancreatic sectors in the earliest contrast-enhanced phases, 2 boluses, each with 2.4 mL of contrast medium, and the technique of "enhancement cancellation" by means of a high-acoustic-intensity flash were used. This technique, during the dynamic observation of the earliest contrast-enhanced phases, almost completely eliminated saturation of the area immediately adjacent to the glandular parenchyma studied, thus obtaining a pure arterial phase there again. The pancreatic head-body sector was examined with the first bolus, and the body-tail area was examined with the second. In this way, it was possible to perform a real-time contrast-enhanced ultrasonographic study of all the accessible glandular sectors, observing the development of the arterial and parenchymal contrast-enhanced phases dynamically. No start delay after contrast material injection was necessary. No disruptive Doppler mode, such as color-stimulated acoustic emission or agent detection imaging, was used to observe the enhancement, which with low-mechanical-index harmonic imaging was immediately visible with the use of second-generation ultrasonographic contrast media. During the second bolus injection, a hypervascular nodule of about 1 cm in diameter, shown as hyperechoic with respect to the adjacent pancreatic parenchyma, was detected and localized at the anterior surface of the body of the pancreas (Fig. 1C
Surgical treatment of the lesion was by laparoscopy, which was performed initially to confirm by direct view the presence of the small nodule on the ventral surface of the pancreatic body (Fig. 2A
Insulinomas are the most common endocrine tumors of the pancreas. They are usually benign tumors, which manifest because of hypersecretion of insulin, responsible for the typical hypoglycemic symptoms. The hypersecretion of insulin is present in the earliest stages of the tumor’s clinical history, which is the reason for their small size at the moment of diagnosis, with the consequent difficulty in detecting and locating them. Ultrasonographic detection of insulinomas is difficult because of the frequent location at the body-tail area of the pancreas and the small sizes of the tumors (55%–70% of insulinomas are <1.5 cm in diameter)2 in generally obese patients. Sensitivity in pancreatic insulinoma identification has reached 79% in the most recent studies, thanks to improvement in ultrasonographic technology in terms of spatial resolution.4 The usually small insulinomas are hypoechoic and well defined without calcifications or necrosis, but these lesions can be isoechoic to the surrounding parenchyma and therefore can be impossible to detect.4–8 Preoperative localization of insulinomas remains extremely difficult even with other imaging modalities, mainly because of the smallness. In the literature, the sensitivity of CT for detecting insulinomas has ranged from 45% to 77%.4,9 Van Hoe et al10 reported higher sensitivity with a bolus-timing technique, showing the insulinomas in about 80% of cases. Conversely, some authors have noted high sensitivity of MR for identifying small pancreatic insulinomas and have proposed this diagnostic modality as the first choice for preoperative imaging.11,12 Nevertheless, according to Ichikawa et al,9 the diagnostic accuracy of MR imaging for detecting pancreatic insulinomas is still similar to that of CT. The literature, however, does agree on the use of invasive methods, such as angiography and venous sampling, as second diagnostic steps and on the important diagnostic contribution of intraoperative ultrasonography, together with manual palpation, in the preoperative detection of lesions.5,11,12 On the basis of the findings in the case reported here, contrast-enhanced ultrasonography could have an important role in detecting pancreatic insulinomas, thus having great potential for identification of pancreatic lesions. The use of a contrast medium during an ultrasound examination can lead to the appearance of small tumors that are unrecognized on a conventional examination because of their hypervascularization. Second-generation ultrasonographic contrast media are characterized by a harmonic response to low-mechanical-index insonation with a greatly enhanced signal, without notable destruction of the microbubbles, thus producing stable and useful contrast-enhanced dynamic phases. In the case reported here, it was possible to observe all the pancreatic sectors dynamically during the contrast-enhanced arterial phase. To our knowledge, there have been no reports in the literature on the use of contrast-enhanced ultrasonography in showing small pancreatic insulinomas. Given the results obtained in the case reported here, however, it would seem feasible to place this new method alongside CT and MR imaging for detecting these small tumors.
Received October 22, 2002, from the Departments of Radiology (M.D., G.M., S.V., C.P.) and Surgery (M.F.), University of Verona, Verona, Italy. Revision requested November 20, 2002. Revised manuscript accepted for publication December 3, 2002.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Top
Introduction
Case Report
