BACKGROUND:

The differentiation of benign from malignant intraductal papillary mucinous tumors (IPMT) is often difficult even by various examination methods. We evaluated the qualitative and quantitative diagnostic ability of contrast-enhanced transabdominal ultrasonography (CE-US), mainly in differentiating benign from malignant tumors in patients with IPMT.

PATIENTS AND METHODS:

There were 21 patients with IPMT who underwent CE-US and endoscopic ultrasonography (EUS). Surgery was performed in all 21 patients. Pathological findings were 4 with carcinoma and 17 with adenoma. CE-US was performed using a contrast agent (Levovist; Tanabe, Osaka, Japan) consisting of galactose microbubbles and a small (0.1%) admixture of palmitic acid, and the following items were evaluated by the following procedure. (1) Two reviewers with experienced sonographic and endosonographic ability evaluated CE-US images before and after contrast enhancement and classified the enhancement effects into three grades. In addition, the presence or absence of enhancement effects by CE-US was compared with that of mural nodules visualized by EUS. (2) In all 21 patients, changes in intensity after contrast enhancement were quantitatively measured using an HDI Lab. HDI Lab was provided by ATL (Philips; Bothell, WA) and these software tools rapidly quantify image characteristics within multiple ROI (regions of interest) and make comparisons between several areas or images. In both the early and late phases, the postenhancement intensity, difference between pre- and postenhancement intensity, and the percentage change ((postenhancement value–preenhancement value)/preenhancement value) were compared between malignant and benign lesions, and the ability of CE-US to differentiate between benign and malignant lesions was evaluated in comparison with the ability of EUS to diagnose the degree of malignancy.

RESULTS:

(1) In both the early and the late phases, both reviewers observed enhancement effects in all 21 patients. And both reviewers observed mural nodules by EUS in all 21 patients. (2) In all 21 patients who underwent resection of IPMT, the intensity increased in both the early and late phases. When the patients with carcinoma were compared with those with adenoma, the postenhancement intensity was significantly higher, and the difference between pre- and postenhancement intensity and the percentage change in the early phase and the late phase was significantly more marked in the carcinoma group (p= 0.019, p= 0.002, p= 0.015, p= 0.012, and p= 0.039, respectively).

CONCLUSION:

CE-US was useful for qualitatively diagnosing tumor lesions in patients with IPMT. Moreover, quantitative changes in intensity can be a parameter for the differential diagnosis of benign and malignant tumors.

INTRODUCTION

Due to recent advances in ultrasonographic equipment and the clinical application of intravenous contrast agents in ultrasonography, contrast-enhanced ultrasonography (CE-US) has been established as a new diagnostic imaging method (^1,2,3,4,5,6). Its usefulness for evaluating hemodynamics in the diagnosis of disorders in various organs including pancreas has been reported. However, most of these studies dealt with the diagnosis of solid tumors and none with pancreatic cystic disease (^7,8).

The usefulness of endoscopic ultrasonography (EUS) for diagnosing pancreatic disease is widely accepted (^9,10). In the diagnosis of IPMT by EUS, the presence of nodules showing papillary elevation in the tumor is considered to be an important parameter of the malignant potential (¹¹). The diagnosis of the presence or absence of nodules is relatively easy by EUS but often difficult by transabdominal ultrasonography. EUS is invasive and causes some degree of burden in patients. If the qualitative diagnosis of IPMT is possible by minimally invasive transabdominal ultrasonography as the primary and follow-up examination, patients with IPMT may receive much benefit (^{12,13,14,15,16}). Therefore, as a new parameter, we evaluated the possible usefulness of the evaluation of septal enhancement effects on CE-US images in diagnosing the degree of the malignancy of IPMT both qualitatively and quantitatively.

PATIENTS AND METHODS

Patients

There were 21 patients (15 males and 6 females) with IPMT who underwent CE-US and EUS between November 1999 and December 2003. Their median age was 67 yr (51–83 yr). In 21 patients, surgical resection was performed, and a definite pathological diagnosis was made (carcinoma, 4 patients; adenoma, 17 patients).

The apparatus used for CE-US was HDI-5000 (ATL (Philips), Bothwell, WA). The apparatus used for EUS was GF-UM200 or GF-UM240 (Olympus, Tokyo, Japan). CE-US was performed with injection of 1 vial of 2.5 g Levovist (Tanabe, Osaka, Japan). Levovist consisted of galactose microbubbles (air microbubbles) and a small (0.1%) admixture of palmitic acid, which after injection into peripheral vein, leads temporarily to enhanced ultrasound echoes. The agent was adjusted to 300 mg/mL in concentration, and was injected intravenously via a 21-gauge needle inserted into the right median antecubital vein as a bolus at a rate of 1 mL/s (8 mL in total volume). The observation procedure was as follows: (1) Early phase: the target lesions were observed continuously by pulse inversion harmonic mode with low mechanical index (MI) value of 0.4 for 60 s after injection of Levovist. (2) Late phase: after a 3-minute suspension of scanning, the lesions were observed by the same mode with a high MI value of 1.3 by a sweeping scan around the lesion. The frame rate was set to a low mode and the focus point was arranged at the distal part of the target lesions. As to the evaluation between plain images and enhanced images, the gain was adjusted in each case to obtain optimal images and was unchanged throughout the sequence for the images of low and high MI values (⁸). To evaluate the image, pairs of still images randomly arranged (plain image vs contrast-enhanced image) were viewed on a computer monitor screen.

The enhancement effects were graded as follows: Grade 0, absence of enhancement effects; Grade 2, definite enhancement effects; and Grade 1 effects between Grades 0 and 2. Grade 1 and 2 effects were regarded as the presence of enhancement effects.

The following items were examined.

1. Two reviewers determined the presence or absence of septal enhancement effects in the tumor in the early and late phases. The presence or absence of enhancement effects by CE-US was compared with the presence or absence of nodules by EUS. In addition, in all 21 patients, the grade classification was compared with histopathological findings.
2. In all patients, a region of interest (ROI) was established by a circle (diameter, 1 mm) in the thickest septal area in the target lesion. The intensity (dB) in the circle was measured three times using an HDI Lab (Advanced Technologies Laboratory, ATL (Philips), Bothell, WA). HDI Lab is the software package that can quantify the image and cineloop display on off-line personal computer for many measurements (¹⁷). In the early and late phases, the intensity before and after Levovist administration was compared in the same area. In the differentiation between benign and malignant lesions, the intensity values in the early and late phases, the differences between the post- and preenhancement (post-pre) intensity, and the percentage change ((post- pre)/pre) were compared between the group with carcinoma and the group with adenoma, and whether differentiation between benign and malignant lesions is possible.

Statistical Analysis

Interobserver variability for the grading by the two reviewers was assessed statistically using the nonweighted statistic. A p-value < 0.05 was considered significant. In 21 patients, various parameters were analyzed by the Mann-Whitney U test.

RESULTS 1

In both the early and late phases, both reviewers observed enhancement effects on the tumor septum in all 21 patients (Table 1).

Table 1 - Coincidence of Reviewers' Grading (Enhancement of Both Early and Late Phases).

Full table

The nonweighted was 0.611: 95% CI [0.268, 0.954], p= 0.006 for early phase, and 0.712: 95% CI [0.411, 1.014], p= 0.001 for late phase, demonstrating the correlation between the grading by reviewers. And both reviewers observed mural nodules by EUS in all 21 patients.

Assuming that the qualitative diagnosis based on septal enhancement effects in the early and late phases observed by CE-US is similar to that based on nodules observed by EUS, in all 21 patients, enhancement grades in the early or late phase determined by reviewer 1 or 2 did not differ between carcinoma and adenoma (Table 1).

RESULTS 2

In all 21 patients who underwent IPMT resection, EUS showed nodules while CE-US showed an increase in intensity in the septal area in both the early and late phases (Figs. 1 and 2).

Figure 1.

Intraductal papillary mucinous adenoma of the pancreas. (A) A pre-enhancement image in the early phase. The mural nodules were not detected on US. Arrowheads indicated the tumor area and the white circle showed the region of interest (ROI) on the septum. (B) A postenhancement image of IPMT in the early phase. The septum of the tumor was increased in echo intensity. (C) A pre-enhancement image in the late phase. (D) A postenhancement image in the late phase. Enhanced echo intensity persisted on the targeted septum. (E) EUS image of the intraductal papillary mucinous adenoma of the pancreas. The mural nodules (arrow) were detected on EUS. (F) The picture of the glass slide demonstrated the multilocular mass with papillary projections (arrows) and the tumor was diagnosed as intraductal papillary mucinous adenoma.

Full figure and legend (127K)

Figure 2.

Intraductal papillary mucinous adenocarcinoma of the pancreas. (A) A pre-enhancement image in the early phase. The mural nodules were not detected on US. Arrowheads indicated the tumor area and the white circle showed the region of interest (ROI) on the thickened septum. (B) A postenhancement image of IPMT in the early phase. The septum of the tumor was increased in echo intensity. (C) A pre-enhancement image in the late phase. (D) A postenhancement image in the late phase. Enhanced echo intensity persisted on the targeted septum. (E) EUS image of the intraductal papillary mucinous adenoma of the pancreas. The mural nodules (arrow) were detected on EUS. (F) The picture of glass slide demonstrated the multilocular mass with papillary projections (arrows) and the tumor was diagnosed as intraductal papillary mucinous adenocarcinoma.

Full figure and legend (131K)

In the differentiation between benign and malignant lesions, the postenhancement intensity was significantly higher, and the difference between post- and pre-enhancement intensity and the percentage change in the early phase and the late phase were significantly more marked in the carcinoma group than in the adenoma group (p= 0.019, p= 0.002, p= 0.015, p= 0.012, p= 0.039), but no difference was observed in the preenhancement intensity in the early phase and the preenhancement intensity and postenhancement intensity in the late phase (p= 0.324, p= 0.473, p= 0.088, respectively) between the two groups (Table 2).

Table 2 - Values of Various Parameters and Statistics.

Full table

DISCUSSION

CE-US, a noninvasive examination method, allowed evaluation of changes in the hemodynamics of pancreatic cystic tumors. In the patients with IPMT, CE-US had septal enhancement effects, and its ability to evaluate the degree of malignancy (adenoma or more advanced lesions) was similar to that of EUS for determining the presence or absence of the nodules. The differentiation between benign and malignant lesions was impossible by qualitative analysis. However, as a result of quantitative analysis using an HDI Lab, the postenhancement intensity in the early phase was higher, and the difference between the post- and pre-enhancement intensity and the percentage change in the early phase as well as the late phase were significantly more marked in the carcinoma group than in the adenoma group. These findings suggest more abundant blood flow in the tumor in the carcinoma group than in the adenoma group. Visualization of nodules is often difficult by US, and nodules are often visualized only as a (thickened) septum on US images. Therefore, it is possible that the presence of nodules might be recognized as the enhanced septum on CE-US images, which might have caused the differences between the adenoma and carcinoma groups even though the septal area was observed similarly on plain US image.

IPMT gradually grows, showing various pathologic conditions from hyperplasia to carcinoma (^14,15). IPMT frequently develops in the elderly and is complicated, often requiring observation of the course. The usefulness of EUS for diagnosing IPMT is established. In follow-up patients, surgery is performed when changes indicating adenoma or more advanced lesions are detected in principle. However, in some patients, observation is continued further due to patients' desire and social background. The diagnosis of malignancy is an absolute indication for surgery. In this study, the results of the evaluation of septal enhancement effects by CE-US were similar to those of the visualization of nodules by EUS. In addition, our quantitative results suggest that septal enhancement effects evaluated using an HDI Lab in CE-US can be a parameter for the differentiation between benign and malignant lesions. In the future, it might be desirable to perform noninvasive CE-US first in the follow-up of patients with IPMT. The summary of the quantitative analyses is as follows. As to prevalues of both early and late phase and post-value in late phase out of all quantitative analyses, there were no significant differences between carcinoma cases and adenoma cases. All other parameters demonstrated significant differences between carcinoma cases and adenoma cases; however, which is the most valuable parameter remains undetermined. The use of an HDI Lab in CE-US allows the evaluation of a quantitative increase in intensity, which is a potential parameter for the differentiation between benign and malignant lesions, but accumulation of additional cases may be necessary. In conclusion, CE-US may be a useful diagnostic procedure in the diagnosis of IPMT.

References

1. Kato, T, Tsukamoto, Y, Naito, Y, et al. Ultrasonographic and endoscopic ultrasonographic angiography in pancreatic mass lesions. Acta Radiologica 1995;36: 381–387.
2. Hirooka, Y, Goto, H, Ito, A, et al. Contrast-enhanced endoscopic ultrasonography in pancreatic diseases: A preliminary study. Am J Gastroenterol 1998;93: 632–635. | Article | PubMed | ChemPort |
3. Kojima, S, Goto, H, Hirooka, Y, et al. Differentiation of benign and malignant lymph nodes with contrast-enhanced echolymphography using endoscopic ultrasound-guided puncture. Hepatogastroenterology 2003;50: 1285–1291.
4. Ugolini, P, Delouche, A, Herment, A, et al. In vitro quantification of flow using continuous infusion of levovist and pairs of harmonic power Doppler images. Ultrasound Med Biol 2001;27: 637–642.
5. Rim, SJ, Leong-Poi, H, Lindner, JR, et al. Quantification of cerebral perfusion with "real-time" contrast-enhanced ultrasound. Circulation 2001;104: 2582–2587. | PubMed | ISI | ChemPort |
6. Orden, MR, Jurvelin, JS, Kirkinen, PP. Kinetics of a US contrast agent in benign and malignant adnexal tumors. Radiology 2003;226: 406–410.
7. Hirooka, Y, Goto, H, Ito, A, et al. Recent advances in us diagnosis of pancreatic cancer. Hepatogastroenterology 2001;48: 916–922.
8. Takeda, K, Goto, H, Hirooka, Y, et al. Contrast-enhanced transabdominal ultrasonography in the diagnosis of pancreatic mass lesions. Acta Radiologica 2003;44: 103–106.
9. Aithal, GP, Chen, RY, Cunningham, JT, et al. Accuracy of EUS for detection of intraductal papillary mucinous tumor of the pancreas. Gastrointest Endosc 2002;56: 701–707. | Article | PubMed |
10. Song, MH, Lee, SK, Kim, MH, et al. EUS in the evaluation of pancreatic cystic lesions. Gastrointest Endosc 2003;57: 891–896. | Article | PubMed |
11. Taki, T, Goto, H, Naitoh, Y, et al. Diagnosis of mucin-producing tumor of the pancreas with an intraductal ultsonographic system. J Ultrasound Med 1997;16: 1–6.
12. Wiesenauer, CA, Shmidt, CM, Cummings, OW, et al. Preoperative predictors of malignancy in pancreatic intraductal papillary mucinous neoplasm. Arch Surg 2003;138: 610–618. | PubMed |
13. Choi, BS, Kim, TK, Kim, AY, et al. Differential diagnosis of benign and malignant intraductal papillary mucinous tumors of the pancreas: MR cholangiopancreatography and MR angiography. Korean J Radiol 2003;4(3):157–162.
14. Sugiyama, M, Izumisato, Y, Abe, N, et al. Predictive factors for malignancy in intraductal papillary-mucinous tumours of the pancreas. Br J Surg 2003;90: 1244–1249. | Article | PubMed | ChemPort |
15. Maire, F, Couvelard, A, Hammel, P, et al. Intraductal papillary mucinous tumors of the pancreas: The preoperative value of cytologic and histopathologic diagnosis. Gastrointest Endosc 2003;58: 701–706. | Article | PubMed | ISI |
16. Waxman, I. Endosonography for differentiating benign from malignant intraductal mucinous tumors of the pancreas: Is the jury out? Am J Gastroenterol 2001;96: 1323–1324.
17. Meairs, S, Daffertshofer, M, Neff, W, et al. Pulse-inversion contrast harmonic imaging: Ultrasonographic assessment of cerebral perfusion. Lancet 2000;355: 550–551.