Insulinomas

Ninety percent of insulinomas are benign and are smaller than 2 cm in diameter. Ninety-nine percent are located in the pancreas. A variety of preoperative imaging modalities for the detection of insulinomas are currently available, such as US, CT, MRI, somatostatin receptor scintigraphy (SRS), and various invasive methods, including endosonography (ES), selective angiography (SA), selective portal venous sampling (PVS), and selective hepatic venous sampling after arterial stimulation (modified Imamura procedure). But those procedures often fail to detect the tumor. On the other hand, there is agreement that skilled surgeons who are experienced in careful and meticulous exploration of the pancreas as well as in the use of intraoperative US (IOUS) can achieve much better results than any of the preoperative methods mentioned, including a combination of most.1,2

Endosonography (ES) is the most sensitive preoperative procedure. It was introduced in the 1980s and provides direct visualization of the pancreas and is able to detect tumors down to 0.3 to 0.5 cm in diameter (Fig. 80-1). An early study by Rosch and colleagues in 19923 identified endocrine tumors by ES in the head of the pancreas in 95% of their patients and in the body and tail in 78% and 60%, respectively (Table 80-1). One year later, Palazzo and coworkers4 underlined its accuracy for localizing small endocrine pancreatic tumors. Thirteen insulinomas less than 15 mm in diameter were imaged by ES, US, and CT. Accuracy for these procedures was 79%, 7%, and 14%, respectively. Since then, we and other have confirmed the results of this method, which is superior to CT, US, MRI, SA, and SRS,5"6 despite the fact that the sensitivity decreases the more left sided the insulinomas are situated. Richards and associates found 83% sensitivity of ES for pancreatic head insulinomas versus 37% for distal pancreatic insulinomas.8 On the basis of these results, ES is the method of choice if one wants to use preoperative imaging. It is mandatory before reoperation or if laparoscopic resection is planned.

CT scanning is probably still the most widely used noninvasive technique for initial localization of insulinomas. It has shown varying results in several extensive studies, with a sensitivity of only 25% to 70%. The most important reason for this is that the sensitivity of CT to localize tumors accurately is dependent on the size and location of the neoplasm.1,6,9"18 Perhaps new techniques such as multislice

FIGURE 80-2. Transverse preoperative ultrasonography (5-MHz linear-array transducer) shows typical hypoechoic insulinoma (15 mm) (TU) within the echogenic parenchyma of the pancreatic head (P). AO = aorta; VC = vena cava; WS = spine.

FIGURE 80-1. Preoperative endosonography shows a typical 22-mm hypoechoic insulinoma (arrows) in the head of the pancreas. D1 and D2 = tumor margins.

FIGURE 80-2. Transverse preoperative ultrasonography (5-MHz linear-array transducer) shows typical hypoechoic insulinoma (15 mm) (TU) within the echogenic parenchyma of the pancreatic head (P). AO = aorta; VC = vena cava; WS = spine.

spiral CT, which is capable of investigating the whole gland in thin sections in 10 to 15 seconds, could improve these numbers (see Table 80-1).

Preoperative US offers a less expensive alternative but is extremely operator dependent. In addition, the sensitivity of US ranges from 0% to 62% and does not exceed that of CT (Fig. 80-2; see Table 80-l).1A9"14'1618

In MRI technique, the improved technology in software, gadolinium-gated studies, magnetic echo delays, and the introduction of oral contrast agents have led to better results in detecting insulinomas.17-19'20 Today, Tl-weighted fat-suppressed images are acquired in arterial phase, portal phase, and equilibrium phase following the administration of intravenous gadolinium.19 However, on the basis of published studies, the sensitivity of MRI is usually 15% to 50% (see Table 80-1).6-9-11-17

In the 1970s, SA was considered a useful localization procedure for detecting insulinomas. Because islet cell tumors are well vascularized, they can be detected by the hypervascular blush if the catheter is placed in the appropriate artery. Different series have shown sensitivity between 35% and 91% (see Table 80-1). However, more recently a decrease in its use has been noted because the excellent sensitivity just mentioned could not be reproduced by many authors. SA also usually identified the larger tumors, which would be relatively easy for experienced surgeons to find at exploration.6,918

Using selective PVS, insulin levels in blood samples obtained from specific sites along the splenic, mesenteric, pancreaticoduodenal, and portal veins are measured. This procedure requires transhepatic catheterization of these veins. It is maximally invasive and can be associated with

TABLE 80-1. Preoperative Localization of Insulinomas: A Literature Review

Technique

TABLE 80-1. Preoperative Localization of Insulinomas: A Literature Review

Technique

US

CT Scan

MRI

SRS

ES

SA

PVS

Authors (Ref.)

n(S%)

n(S%)

n (S%)

n(S%)

n (S%)

n (S%)

n(S%)

Vinik et al,9

6(0)

25 (16)

4(0)

35 (42)

32 (81)

Galiber et al.!0

28 (61)

23 (30)

26 (54)

Doherty et al.11

23 (26)

23 (17)

8(25)

26 (35)

22 (77)

Fraker et al.12

32 (22)

14 (43)

19 (84)

7 (100)

Daggett et al.1

11 (45)

8 (37)

_

18 (56)

8 (25)

Rothmund et al.'3

142 (39)

246 (33)

305 (61)

80 (89)

Bottger et al.14

21 (62)

15 (73)

30 (67)

13 (77)

Pasieka et al.16

35 (26)

18 (44)

36 (94)

Rosch et al.3

31 (81)

Palazzo et al.4

13(8)

13(15)

13 (77)

Kuzin et aI.1B

78 (30)

38 (24)

118 (56)

Boukhman et a!.17

64 (50)

64 (24)

64 (40)

64 (47)

64 (55)

Hashimoto and Walsh13

6(0)

20 (15)

3(33)

10 (48)

5 (50)

Fendrich et al.6

33 (33)

27 (33)

13 (15)

14 (0)

23 (65)

8 (37)

2 (1D0)

(S%) = sensitivity rate; US » abdominal ultrasonography; CT scan = computed tomography; MRI = magnetic resonance imaging; SRS = somatostatin receptor scintigraphy; ES - endosonography; SA » selective angiography; PVS selective portal venovis sampling.

(S%) = sensitivity rate; US » abdominal ultrasonography; CT scan = computed tomography; MRI = magnetic resonance imaging; SRS = somatostatin receptor scintigraphy; ES - endosonography; SA » selective angiography; PVS selective portal venovis sampling.

serious complications. Significantly increased hormone concentrations from one area compared with others are considered a positive study result. Although PVS could reach sensitivity rates of 80% to 100%6-911-'5-21 (see Table 80-1), the procedure does not really localize but only helps to regionalize the tumor to the part of the pancreas drained by a particular vein.22

SRS was believed to be a promising method after the initial data were published. Large numbers of somatostatin receptors (SS-Rs) are found on most endocrine pancreatic tumors. At least five different human SS-R subtypes have been cloned. Octreotide binds with high affinity to SS-R subtype 2 (sst2) and sst5.23 The efficacy of SRS using this agent in 350 patients with proven endocrine tumors was evaluated in a European multicenter trial.24 The highest success rates were observed with glucagonomas (100%), vasoactive intestinal polypeptide-secreting tumors (88%), gastrinomas (73%), and nonfunctioning islet cell tumors (82%). Insulinomas were detected in only 46% to 63% of cases due to the low incidence of sst2 on insulinoma cells. Therefore, the method is already losing favor (see Table 80-1).

In this context, a modification of the Imamura method is also worth mentioning. In this procedure, known as arterial stimulation and venous sampling, calcium gluconate is injected into various gastroduodenal and splenic arteries. After the injection, blood is obtained for insulin assay from the hepatic veins.25 Doppmann and others demonstrated high (88% to 100%) sensitivity rates25"27 of this method, which is less painful, is less difficult for the interventional radiologist to perform, and is associated with fewer complications than transhepatic catheterization of the pancreatic veins.

Daggett and colleagues1 were the first to show that intraoperative exploration of the pancreas might be the best method to localize insulinomas. In their survey, 29 patients underwent laparotomy for suspected insulinoma. The tumors were correctly localized before operation in 13% by CT, in 18% by US, in 25% by selective PVS, and in 50% by SA. On the other hand, the tumors were detected and resected in 27 of these 29 patients at surgical exploration of the pancreas. Since then, many studies2'10'13,14,1718,21 confirmed these results, and our survey6 of 40 patients with insulinoma has shown that the tumor was correctly localized before operation in 65% by ES, 37% by SA, 33% by CT and US, 15% by MRI, and 0% by SRS. On the other hand, all tumors were identified and resected using IOUS and meticulous palpation of the pancreas after extensive mobilization of the gland.

Recommendation

We recommend US or CT scan as the only preoperative test before primary operations, not to find the tumor but to exclude metastases of possibly malignant insulinomas, which are usually found in the liver. The patients should then undergo laparotomy that includes meticulous surgical exploration, including an extended Kocher maneuver to be able to palpate the head of the pancreas and mobilization of the body and tail from the retroperitoneum (including the spleen if necessary) to examine the distal pancreas carefully and completely. IOUS should then be used to confirm the presence of tumor or to find nonpalpable lesions and also to determine the relation of the tumor to the pancreatic duct. This should be performed by an experienced surgeon who is familiar with IOUS or asks an ultrasonographer to participate. If one desires a preoperative method besides US or CT, ES is the procedure of choice, but it can be omitted according to our experience. For patients requiring reoperation or for patients in whom laparoscopic resection is planned, ES is recommended as the second procedure after US or CT. When no lesion is identified and one can rely on the biochemical tests for diagnosis, laparotomy should follow.

Diabetes 2

Diabetes 2

Diabetes is a disease that affects the way your body uses food. Normally, your body converts sugars, starches and other foods into a form of sugar called glucose. Your body uses glucose for fuel. The cells receive the glucose through the bloodstream. They then use insulin a hormone made by the pancreas to absorb the glucose, convert it into energy, and either use it or store it for later use. Learn more...

Get My Free Ebook


Post a comment