The number and influence of animal studies have been limited. However, advances in the clinical application of EUS depend on refining image interpretations. The study of EUS in animals should improve interpretation of sonographic images through correlation of histology of disease to sonography.

In this chapter, optimal methods for performing EUS and factors such as ultrasound image artifacts, which contribute to accurate EUS interpretation, will be reviewed and discussed. The EUS technique can then be applied to understanding mechanisms inherent to GI tissue pathologies. The applicability of EUS to clinical problems will also be addressed.

A reflecting mirror converts the sector scan from perpendicular to parallel to the scope. Most echoendoscopes currently have a side or oblique angle of view with the exception of the echocolonoscope which is forward viewing. The echocolonoscope has an accessory channel with the optical components resulting in a 40' wedge cut from the full 360' EUS image. The echoendoscope can be passed into the distal duodenum, and the caecurn can be reached with the echocolonoscope.

As with all endoscopic techniques, passing the instrument through the lumen is relatively easy. The difficulty arises in passing the scope to the target area, identifying and examining the target, and then interpreting the images and artifacts.

EUS is performed in a fasting patient. Pharyngeal anaesthesia and intravenous sedation are generally used, but are not required. Endoscopy with a standard endoscope is performed initially because the echoendoscope has limited capabilities as an endoscope. The rigid tip and the oblique viewing optics of the echoendoscope make examination of the mucosa and the lumen more difficult than with standard endoscopes, and similar to that with a large diameter side-viewing endoscope. Initial standard endoscopy allows for assessment of endoscopic anatomy and detection of areas such as strictures that may be at risk for a complication.

During EUS scanning, the transducer is guided through the GI lumen and placed directly adjacent to the area of interest. Bones, adipose tissue and air-filled structures, which limit the sound wave imaging clarity of extracorporeal sonography, are avoided. The close proximity of the ultrasonic source to the organs imaged permits the use of high frequency, high resolution sound waves that have too short a penetration depth to be used in transcutaneous sonography. Current EUS instruments use the higher sound wave frequencies of 7.5 and 12 MHz, as compared with 3.5 MHz for transcutaneous US and 5 MHz for blind rectal and oesophageal probes.

As with all sonography, the sound wave imaging plane can be oriented at any angle to bring the lesion into optimal focus simply by turning the probe. The operator is not limited to parallel sections, usually 1cm apart, as with computed tomography (CT). When the echoendoscope is in the target area, intraluminal gas is aspirated to increase contact with the GI wall, thus maximizing available acoustic windows. The area to be imaged is brought into focus by filling a thin Latex rubber balloon (Olympus Optical Co. Ltd) surrounding the transducer at the tip of the echoendoscope with approximately 15 ml water. The water-filled balloon serves as a non-attenuating, acoustic coupling medium between the transducer and the surface of the GI wall. If the tar,et area is in the wall itself, water (up to 1000 ml) can be placed directly into the lumen for optimal acoustic coupling. The scanning plane is then manoeuvered, using scope controls and turning, pushing or pulling the scope shaft to orient the lesion into optimal position and focus. The technique will vary according to the indications arid objectives of the examination.

1. Equipment
   1. Malfunction or improper operation
   2. Improper calibration
   3. Imaged object out of frequecy's focal range
   4. Improper adjustment of electreical controls
2. Acoustic presentation of the image
   1. Assumptions made by the sonographic instrument to produce an image
   2. Physical properties of the sound beam
   3. Acoustic properties of the tissue
   4. Interaction of sound and tissue
3. Characteristics of the tissue or medium
4. EUS technique
   1. Improper operation of the transducer resulting in:
      1. Non-perpenducular scanning
      2. Object compression artifacts
      3. Insufficient contact at appropriate anatomical acoustic window
      4. Misinterpretation of anatomy
   2. Improper focal length adjustment altering acoustic presentation of the image

Absorption, reflection, refraction and scatter are behaviours of sound waves as they propagate through and interact with tissue. Through absorption the mechanical energy of the sound pulse is converted to heat. Reflection and refraction refer to the portion of sound that returns or emerges from a boundary or interface of a medium. Scattering occurs because of diffusion or redirection of sound in various angles when it encounters a particle suspension or a rough surface, resulting in sound energy not returning to the transducer and, therefore, loss of its detection by the transducer. Acoustic impedance (resistance) is the product of wave velocity through the medium and density of the medium. All biological tissue and media have inherent acoustical properties and impedance. As sound travels through a medium, it looses energy through interactions of sound and tissue, and becomes attenuated. Higher frequency sound waves are subject to a greater degree of attenuation. Spatial resolution is also related to frequency. The higher the frequency, the better the spatial resolution. Although high frequency sound waves can penetrate deeper levels, images will be out of focus when attenuation overcomes the gain in spatial resolution.

Reflection is an important property of sound waves required for image formation in US. Sound reflection is maximized when a high amplitude beam strikes a soft tissue-gas or fluid-gas interface at a perpendicular angle. When the acoustic impedance of two adjacent tissues is different, sound striking the tissue interface is reflected, producing a sound wave echo which returns to the transducer. The reflected sound wave is translated to the screen, placing the interface at a specific point in the image that correlates with the distance from the transducer to the interface in the tissue. The greater the difference in tissue impedance, the stronger the amount of reflection. Reflection is also intensified with greater amplitude of the incident beam.

There are certain artifacts which relate to equipment malfunction. These occur when air, water or dust gain access to the oil-immersed-transducer housing, the transducer, or the electrical connections. These types of artifacts are generally easy to detect as they will be present on the screen even when the instrument is not being used for scanning.

Because of attenuation, the optimal focal range of an instrument will vary with frequency. For higher frequencies, the focal length is smaller and closer to the transducer; for lower frequencies, the focal length is further away from the transducer. The optimal focal range for 7.25 MHz is 1-4 cm and for 12 MHz, 2-20 mm. Objects that are not in the focal range of the Sound beam cannot be consistently and accurately deciphered (Fig. 13.2a and b). Adjustment of image intensity with amplification (gain) or using adjustments in contrast, focal zone and depth can compensate for some of these factors in producing an accurate image. However, improper machine settings as well as improper adjustments of equipment controls during, scanning can cause artifacts.

Figure 13.2a

Figure 13.2b

Acoustic artifacts

The US instrument generates, records and processes complex signals using assumptions that are simple and consistent, but flawed in all but ideal situations. Artifacts result when significant violations of these assumptions occur. These assumed principles include: Sound always travels in straight lines; only the properties of the imaged object directly determine the intensity of returning echoes; distance is directly proportional to the time it takes for an echo to make a round trip and return along the propagation path to the surface of the transducer.

Reverberation occurs as sound strikes a subjacent interface and is reflected many times between the interface and the transducer surface. When these multiple reflections are strong enough, they are detected by the receiver and given a spatial alignment that is a multiple of the depth of the original reflective interface. These artifactual echoes occur especially within soft tissue or fluid structures, and sianificant1v alter the echo texture of the object. Reverberations can simulate disease, such as a pseudomass or thrombus in a vessel, when the additional 'non-real' reflectors are placed on the image. The water-filled balloon around the EUS transducer can also be a cause of reverberation artifacts.

Short-path reverberations cause comet tail or ring down artifacts, seen behind a small but intense reflector such as air, metal, plastic or calcified objects. They appear as a series of closely spaced echoes that trail off in intensity as distance from the object increases. The short path is probably produced by microbubbles or crystalline structures that set up reverberation chambers.

Multipath artifacts occur when the reflected sound beam maintains its intensity and coherence at flat and smooth interfaces. Acoustic noise is produced by back scattering of many secondary sound waves from surrounding tissue which reflect again off the smooth interface and return signals to the transducer surface. The result is the visible acoustic noise or dirty shadowing behind a reflecting surface such as a large, smooth calcification or gas pocket. However, the coherence of the reflected beam depends on the diffractive nature of the presenting Surface. If the insonified surface is rough and/or has a small radius of curvature, the back-scattered beam will be diffuse, producing phase incoherence in the return beam. Since absorption, not reflection, of sound wave is the dominant process, phase cancellations and loss of signal occur, which cause clean shadowing.

Mirror image artifacts occur when sound takes longer, more indirect path from the primary interface to a secondary interface before finally returning to the transducer surface. The processor assumes a straight line path and places a phantom lesion at a location deep to the primary reflector due to misregistration of some secondary reflectors. Mirror image artifacts are commonly found around the diaphragm, pleura and bowel.. Similar phenomena are seen in colour Doppler.

Side lobe artifacts (from single element transducers) and grating lobe artifacts (from arrays) result from several low-intensity side lobe sound beams around the main ultrasound beam. These side beams can interact with reflectors and present sound back to the transducer face causing objects to be displayed incorrectly in a lateral position. The instrument, believing the integrity of a single main beam, assigns these side echoes a fictitious position within the path of the main beam. Side lobe signals are most significant at highly reflective interfaces and cause the true echo texture of the imaged object to be altered by low level echoes.

Focal zone banding artifacts occur because brighter shades on the grey scale are always assigned to higher amplitude echoes. However, a sound bearn varies in amplitude along its propagation path, resulting in focal zones of increased intensity. Electronic focusing can therefore create focal zone bands of alternating high and low intensity in an organ that is actual1v homogeneous. A pseudhypoechoic mass can be created in an organ through such banding.

Flat artifact is seen with colour Doppler and has not played a significant role in EUS. It occurs when colour is suppressed where grey-scale echoes are present, but assigned to anechoic or hypoechoic areas.

Spatial resolution limitations call result in artifacts. Objects that are separated by a small distance call merge on the screen if the pulse length is not short enough to distinguish two closely spaced points. Axial resolution is superior to lateral resolution because pulse length is normally much less than pulse width. Axial resolution generally improves ,with higher frequencies. Lateral resolution call range up to 3 cm, whereas axial resolution is usually no more than 2-3 mm. Therefore, small objects may appear larger or thicker when reflectors are parallel to the bearn compared to those that are encountered perpendicular to the beam. Measurements are best made in the axial direction as often as possible.

Image speckle, an interference pattern close to the transducer, causes a parenchymal echopattern, but also image degradation. This acoustic noise is produced by the constructive and destructive interference of rotating echoes from a scatter distribution. Speckle call be reduced by photographic averaging and deconvolution resulting in, improved images.

Slice thickness artifact, also termed section thickness, off-axis or beam-width artifact. results in the melding of the image of different tissues that may not belong together. A focused US pulse has a finite width in the direction perpendicular to the scan plane. The beam can therefore sonographically locate more than one tissue at the same location of a scan plane. The two simultaneously imaged tissues produce all image that has a combined echo texture.

Speed propagation artifact occurs because sound propagates through different body tissues or media at different velocities. Sound travels more slowly through soft tissue than fluid, and even more slowly through fat. As a result, sound that must traverse fatty tissue completes its round trip back to the transducer in a longer period of time. However, to produce an image, US equipment assumes an average tissue velocity of 1540 cm/s. Based on round-trip time and inaccurate depth assignment is made to reflectors posterior to tissue such as fat, fluid and cartilage which have acoustical velocities significantly different from the average.

Through-transmission artifacts occur when sound travels through a medium with low attenuation properties before reaching the imaged object. Scanning through fluid will cause objects to be more echogenic than usual. This is particularly relevant or EUS when cysts or peritoneal fluid are encountered around an area of disease.

Refraction occurs when the sound beam at an oblique angle of incidence strikes a boundary between two media which conduct sound at different velocities. The transmitted part of the sound wave can be bent. The proportion of refraction and reflection that occurs depends on the angle of incidence. Complete bending of the beam can occur when the beam is parallel to the interface, obscuring the image of tissue posterior to the interface. Depending on the direction of the sound beam through the two adjacent media, sound waves will converge or diverge,

Posterior shadowing artifacts occur when sound wave penetration of a structure is very limited or absent. When impedance of two tissues differs to Such a degree that a near complete reflection of the Sound beam is produced, a dark shadow is noted where there are no echoes posterior to the interface. Shadowing also occurs at two apposed transitional zones in the same tissue. Refraction can also cause shadowing from the edge of a curved object, typically at the interface of calcium/soft tissue or gas/fluid.

Enhancement artifacts are due to contrasting acoustic impedance in adjacent structures. Posterior enhancement is a deceptive brightness behind a dark area. The higher amplitude of these echoes compared to those of equivalent adjacent tissue arises posterior to a contrastingly low attenuating area, such as a fluid structure. This sound transmission artifact can be used to distinguish cystic from solid structures. Edge enhancement is an infrequently seen artifact which can occur due to refraction at a curved edge. Focal enhancement can occur in the focal region of the transducer.

Phase cancellation occurs when reflected wave segments are out of phase and cancel each other in summation. When a coherent sound wave crosses boundaries between media that conduct sound at different velocities, the adjacent waveforms become distorted and out of phase. The destructive interference produces an area of cancellation resulting in a black streak or absence of signal on the screen.

Attenuation is a loss of sound wave intensity as a consequence of the behaviours of sound as it interacts with matter. The greatest loss is due to absorption. The higher the frequency, the greater the absorption of sound energy. The degree of attenuation is related to wave frequency and the tissue type and shape. Hypoechoic regions, which could be misinterpreted as masses, can occur behind certain tissues that significantly attenuate the sound beam. Because of the high frequency used with EUS, attenuation artifacts must always be considered. Accuracy in imaging the deep parts of soft tissue structures can be limited by shadows produced when a sound beam does not traverse the structure.

Anisotrophy artifacts result from changes in tissue plane orientation relative to the main second beam axis due to the modulating shape of an imaged organ. Tissue echogenicity changes as sound waves are transmitted at different angles through the same structures due to alterations of sound wave behaviour at the tissue reflectors.

Perivascular colour artifact occurs when intravascular blood flow becomes turbulent resulting in vibrations within disturbed tissue surrounding the blood vessel. When the tissue is examined with colour Doppler, the motion of this vibration will be colour-encoded producing a sonographic equivalent of a soft tissue thrill.

The composition of different tissues and organs determines their ultrasound image characteristics. Acoustic impedance of soft tissues can differ by as Much as 22%. The presence and uniformity of distribution of different tissue components in a soft tissue structure alters its US image. Changes in tissue density and uniformity due to disease can also alter sound wave/tissue interactions. Inherent complexity of tissue properties will be the major consideration for in vitro studies. For in vivo, and human studies, knowledge of anatomical relationships is also critical to proper EUS interpretation.

Important sources of artifacts and errors that arise from improper instrument placement are non-perpendicular scanning, object compression, insufficient contact at the appropriate anatomical acoustic window, and misinterpretation of anatomy. When the incident sound wave encounters a reflector at an angle, the reflected wave will emerge at an equal angle. Part of the incident wave may not reflect, but may be transmitted into the next tissue or medium. The angle of scanning will affect the amplitude of reflected sonographic echoes and the corresponding degree of refraction. Transducer position can affect these properties of sound waves and can produce artifacts.

The most important source of EUS transducer positioning artifact is oblique or tangential scanning, which can result in the appearance of pseudotumours or indistinct margins between parenchyma and vessels through acoustic artifacts such as reverberation, shadowing and enhancement. However, the most common distortion of non-perpendicular scanning is the appearance of a widening or thickening of the layers of the intestinal wall. The best measurements are obtained with the transducer at right angles to the target so that the sound beam is perpendicular to any boundaries between two tissues or media. In performing EUS, one is always trying to optimize image clarity by balancing the highest frequency to obtain the best spatial resolution with a frequency that penetrates deep enough to view the target area. The higher the frequency, the less divergence occurs, and the easier it is to focus the narrowed beam.

The amount of water placed in the balloon surrounding the transducer affects the way the image is generated and can alter the size of the field of optimal focus. Changing the amount of water in the balloon can also be used to move the focal point. However, changing the angle of the echoendoscope and the amount of water in the balloon can easily compress structures so that distortion occurs. Superficial layers of the wall are more sensitive to this distortion than other parts of the intestinal tract or surrounding structures.

Repeat scanning for the same as well as different angles during introduction and withdrawal of the instrument while varying the focal length with the balloon will limit misinterpretation of an EUS image.

The effects of artifacts can be minimized by scanning from different angles, using the narrowest sound beam possible, focusing properly on the target area, avoiding scanning at edges of objects, recognizing secondary images, and adjusting equipment settings (usually the gain). Misinterpretation of anatomy often results from improper transducer location and/or orientation. Misinterpretation also results if surrounding anatomy and structures of the area being scanned are not considered. This is particularly true at the distal oesophagus where the diaphragm complicates interpretation, especially if a hiatal hernia is present.

Animal studies have not been as numerous as human studies. Since future improvements of the clinical application of EUS will depend on advances in image interpretation, useful information could potentially be gained through animal studies. Models could be developed to improve correlations of sonographic findings to histology. The investigation of various agents which enhance or change sound wave characteristics of tissues will assist in differentiating normal and pathological states.

Pathology in GI organs can be imaged with EUS using anatomical landmarks for orientation. In patients without gastric surgery, all landmarks can be located in at least 90% of cases. However, scanning all organs completely would take more than an hour. Thus, in any given examination, certain unrelated structures or areas need not be recorded, if the focus is a specific anatomical site.

		EUS	CT
Oesophagus	T stage	85	60
	N stage	80	55
	M stage	70	70
	Resectability	80	55
Gastric	T stage	80	40
	N stage	75	50
	M stage	75	75
	Resectability	80	70
Pancreas	T stage	90	50
	N stage	75	50
	M stage	65	65
	Resectability	80	40
Biliary System	T stage	85	45
	N stage	60	50
	M stage	85	85
	Resectability	80	50
Rectum	T stage	85	70
	N stage	80	55
	M stage	-	75

EUS provides detailed images of the GI tract. Clinical applications continue to expand. Proper use of equipment and understanding sound wave properties will minimize errors of the method. Appreciation of how artifacts produce certain changes in tissues improves the operator's facility not only to distinguish the true image, but also to use the artifact to interpret pathology and make a diagnosis. Shadowing and enhancement artifacts are frequently useful in this regard. Section thickness and reverberation artifacts are generally more difficult to interpret and use constructively. The sonographer can be seriously misled by artifacts, which if properly recognized can be used to reveal and inform.

Acknowledgement

The author is grateful to Laurel Kiefer for editorial and graphic assistance.

1. Wild JJ, Reid JM. Diagnostic use of ultrasound. BrJ Phys Med 1956; 19: 248-257.

2. Lutz H,Rosch W. Transgastroscopic ultrasonography. Endoscopy 1976; 8: 203-205.

3. Sasai T. Development of ultrasonic endoscope. In: Kawai K (Ed) Endoscopic Ultrasonography in Gastroenterology Igaku-Shoin Ltd, Tokyo, 1988, pp. 18-34.

4. Snady H. The role of endoscopic ultrasonography in diagnosis, staging and outcomes of gastrointestinal disease. Gastroenterologist 1994; 10: 91 -110.

5. Snady H. Technical and interpretive pitfalls in initial experience with endoscopic ultrasonography for upper gastrointestinal disease. Lessons learned. Gastroenterology 989; 96: A480.

6. Rosch T. Endoscopic ultrasonography artifacts and problems of interpretation. Gastrointest Endosc 1996; 43: S10-S12.

7. Tytgat G NJ, Tio TL (Eds). Endoscopic Ultrasonography: Proceedings of the 4th International Symposium on Endoscopic Ultrasonography. Scand J Gastroenterol 1986; 21(suppl 123): 1-172.

8. Kawai K (Ed). Endoscopic Ultrasonograpliy in Gastroenterology Igaku-Shoin, Tokyo, 1988.

9. Snady H. Endoscopic ultrasonography: an effective new tool for diagnosing gastrointestinal tumors. Oncology 1992; 6: 63-74.

10. Rosch T, Classen M. Gastroenterological Endosonography (Textbook and Atlas). Thieme Medical Publishers, New York, 1992.

11. Lightdale C (Ed). Endoscopic ultrasonography. In: Sivak M Jr (Ed) Gastrointest Endosc Clinics N Am 1992; 2: 557-749.

12. Rosch T (Ed). Endoscopic ultrasonography: State of the art - 1995. Part 1. In: Sivak M Jr (Ed) Gastrointest Endosc Clinics N Am 1995; 5: 475-698.

13. Rosch T (Ed) Endoscopic ultrasonography: State of the art - 1995. Part II. In: Sivak M Jr (Ed) Gastrointest Endosc Clinics N Am 1995; 5: 699-898.

14. Gillard V, Mainguet P, Vicari F, Florent Ch (Eds). 3rd Endoscopic Ultrasonography Belgian Meeting. Acta Endoscopia 1995; 25: 407-556.

15. Erickson R, Chang K. Normal anatomy, part 1. In: Training for Endosonography - An Interactive Learning Tool. CD-ROM, Olympus America and Astra Merck Inc, 1996.

16. Caletti G, Gerrari A, Barbara L. Normal endosonographic anatomy of the esophagus and stomach. In: Lightdale C (Ed) Endoscopic ultrasonography. Gastrointest Endosc Clinics N Am 1992; 2: 601-614.

17. Snady H. Encloscopic ultrasonography images of the normal retroperitoneum. In: Lightdale C (Ed) Endoscopic ultrasonography. Gastrointest Endosc Clinics N Am 1992; 2: 637-655.

18. Dancygier H. Endosonographic evaluation of biliary tract disease. In: Lightdale C (Ed) Endoscopic ultrasonography. Gastrointest Endosc Clinics N Am 1992; 2: 697-714.

19. Wiersema MJ, Hawes RJ. Normal colorectal anatomy and benign colon lesions. In: Lightdale C (Ed) Endoscopic ultrasonography. Gastrointest Endosc Clinics N Am 1992; 2: 715-728.

20. Vilmann P. Endoscopic ultrasonography-guided fine needle aspiration biopsy of lymph nodes. Gastrointest Endosc 1996; 43: S24-S29.

21. Silverstain FE, Martin RW, Kimmey MB, Jiranek GC, Francklin DW, Proctor A. Experimental evaluation of an endoscopic ultrasound probe: in vitro and in vivo canine studies. Gastroenterology 1989; 96: 1058-1062.

22. Taniguchi D, Martin R, Trowsers E, Silverstein F. Simultaneous M-mode echoesophagram and manometry in sheep esophagus. Gastrointest Endosc 1995; 41: 582-586.

23. Gress F, Park K, Sangvi N, Kopecky K, Hawes R. A comparison study of high frequency ultrasound imaging of the canine GI tract at 20, 30 and 50 MHz frequencies. Gastrointest Endosc 1996; 43: s54.

24. Takemoto T, Yanai H, Tada M et al. Application of ultrasonic probes prior to endoscopic resection of early gastric cancer. Endoscopy 1992; 24(suppl 1): 329-333.

25. Bartram CI. Anal sphincter disorders. Gastrointest Endosc 1996; 43: S32-S34.

26. Yasdua K. Ultrasonic probes For pancreaticobiliarv strictures. Gastrointest Endosc 1996; 43: S35-S37.

27. Odegaard S, Kimmey M. Location of the Muscularis mucosa on high frequency gastrointestinal ultrasound images. Eur J Ultrasound 1994; 1: 39-50.

28. Wiersema M, Wiersema L. High-resolution 25 Megahertz ultrasonography of the gastrointestinal wall: histologic correlates. Gastrointest Endosc 1993; 39: 499-504.

29. Tio TL, Tytgat GNJ. Endoscopic ultrasonography in analyzing per-intestinal lymph node abnormality. Preliminary results in vivo and in vitro. Scand J Gastroenterol 1986; 21(suppl 123): 158-163.

30. Behahrs 0, Henson D, Hutter R (Eds). Manual for Staging of Cancer, 3rd edn. American joint Committee on Cancer. JB Lippincott Co, Philadelphia, 1988.

31. Hermanek P, Sobin L (Eds). International Union Against Cancer (UICC): TNM classification of malignant tumours 4th edn. Springer-Verlag, Berlin, 1987.

32. Denoix PF, Six annees d'enquete permanente cancer. Bull Inst Natl Hyg 1944; 1: 1-69.

33. Denoix PF. De L’importance d'une nomenclature unifiee dans l'etude du cancer. Rev. Med Franc 1947; 36: 1124-1128.

34. Hutter RV. At last - Worldwide agreement on the staging of cancer. Arch Surg 1987; 122: 1235-1239.

35. Tio L. The TNM staging system. Gastrointest Endosc 1996; 43: S 19-24.

36. Snady H, Bruckner H, Siegel J, Cooperman A, Neff R, Kiefer L. Endoscopic ultrasonographic criteria of vascular invasion by potentially resectable pancreatic tumors. Gastrointest Endosc 1994; 40: 326-333.

37. Palazzo L, Burtin P. Interobserver variation in tumor staging. In: Rosch T (Ed) Endoscopic Ultrasonography: State of the art - 1995. Part 1. In Sivak M Jr (Ed) Gastrointest Endosc Clinics N Am 1995; 5: 559-568.

38. Roubein L. Inter-observer variability in endoscopic ultrasonography: a prospective study. Acta Eitdoscopica 1995; 25: 549-550.

39. Caletti GC, Brocchi E, Ferrari A et al. Guillotine needle biopsy as a supplement to endosonography in the diagnosis of gastric submucosal tumors. Endoscopy 1991; 23: 251-254.

40. Wiersema M, Hawes R, Tao L-C et al. Endoscopic ultrasonography as an adjunct to fine needle aspiration cytology of the upper and lower gastrointestinal tract. Gastrointest Endosc 1992; 38: 35-39.

42. Kimmey M, Martin R. Fundamentals of endosonography. In: Lightdale C (Ed) Endoscopic ultrasonography. Gastrointest Endosc Clinics N Am 1992; 2: 557-573.

43. Kremkau F, Taylor K. Artifacts in ultrasound imaging. J Ultrasound Med 1986; 5: 227-237.

44. Scanlan KA. Sonographic artifacts and their origins. Am J Radiol 1991; 156: 1267-1272.

45. Rubin J, Adler R, Bude R, Fowlkes J, Carson P. Clean and dirty shadowing at US: a reappraisal. Radiology 1991; 181: 231-236.

46. Kliewer MA, Hertzberg BX, George PY, McDonald JW, Bowie JD, Carroll BA. Acoustic shadowing from uterine leiomyomas: sonographic-pathologic correlation. Radiology 1995; 196: 99-102.

48. Kimmey MB, Martin RW, Haggitt RC, Wang Y, Franklin DW, Silverstein FE. Histologic correlates of gastrointestinal ultrasound images. Gastroenterology 1989; 96: 433-441.

49. Aibe T, Fuji T, Okita K, Takemoto T. A fundamental study of the normal layer structure of the gastrointestinal wall visualized by endoscopic ultrasonography. Scand J Gastroenterol 1986; 21(supp123): 6-15.

50. Tio TL, Tytgat GNJ. Endoscopic ultrasonography of normal and pathologic upper gastrointestinal wall structure. Scand J Gastroenterol 1986; 21(suppl 123):27-33.

51. Boscaini M, Montori A. Transrectal ultrasonography: Interpretation of normal intestinal wall structure for the preoperative staging of rectal cancer. Scand J Gastroenterol 1986; 21(suppl 123): 87-98.

52. Bolondi L, Caletti G, Casanova P, Villanacci V, Grigioni W, Labo G. Problems and variations in the interpretation of the ultrasound feature of the normal upper and lower GI tract wall. Scand J Gastroenterol 1986; 21(suppl 123): 16-26.

53. Silverstein F, Kimmey M, Martin R et al. Ultrasound and the intestinal wall: Experimental methods. Scand J Gastroenterol 1986; 21(suppl 123): 34-40.

54. Heinz A, Mildenberger P, Georg M, Garcia A, Junginger Th. In vitro studies of lymph node analysis. In: Rosch T (Ed) Endoscopic ultrasonography: State of the art - 1995. Part 1. Gastrointest Endosc Clin N Am 1995; 5: 577-586.

59. Mattrey RF, Strich G, Shelton RE et al. Perfluorochemicals as US contrast agents for tumor imaging and hepatosplenography: preliminary clinical results. Radiology 1987; 163: 339-343.

60. Fritzsch T, Hilmann J, Kampfe M, Muller N, Schobel C, Siegert J. SHU508, a transpulmonary echocontrast agent: initial experience. Invest Radiol 1990; 25: S160-S161.

61. Gallez B, Demeure R, Debuyst R et al. Evaluation of nonionic niroxyl lipids as potential organ specific contrast agents for magnetic resonance imaging. Mag Res Imag 1992; 10: 445-455.

62. Bhutani MS, Hoffman BJ, Hawes RH. A swine model for endosonographic pancreatic imaging (Abstract). Gastrointest Endosc 1996; 43: s53.

65. Taniguchi D, Martin R, Trowers E, Silverstein F. Simultaneous M-mode echoesophagram. and manometry in sheep esophagus. Gastrointest Endosc 1995; 41: 582-586.

66. Lees WR. Endoscopic ultrasonography of chronic pancreatitis and pancreatic pseudocysts. Scand J Gastroenterol 1986; 21(suppl 123): 123-129.