- Email: [email protected]
⁎3436 Enhanced magnification endoscopy: a new technique to identify specialized intestinal metaplasia in barrett's esophagus.
*3433 GALLBLADDER WALL LAYER DISPLAYED BY MR ENDOSCOPY. Fumiaki Kitahara, Tadashi Sato, Yuichiro Kojima, Atsuro Morozumi, Masayuki A. Fujino, Tomoaki Ichikawa, Tsutomu Araki, Yamanashi Med Univ, Japan. Introduction: Imaging of gallbladder wall layers is almost performed by endoscopic ultrasonography. Standard magnetic resonance (MR) imaging of gallbladder by body and surface coils often inadequately depict extension into the gallbladder wall. Endocavitary coils have been developed to help solve this problem. Aims: To investigate MR endoscopic imaging of gallbladder wall layer compared to histopathologic appearance. Methods: We used an MR endoscope, type XGIF-MR30, developed by the Olympus Optical Co. Ltd. (Tokyo, Japan), with a 3-cm-long receive-only coil embedded in its tip.All imaging is currently performed on a 1.5-T system (Signa 5.6; GE Yokogawa Medical Systems, Tokyo, Japan). Patients were sedated intravenously in a standard fashion, and anticholinergic agent is administered intravenously to minimize gut motility and spasm. The tip of the MR endoscope was placed in the duodenal bulb. Then, axial MR examinations were performed in the following sequences: T1-weighted imaging, fast-spin-echo T2-weighted imaging, MRCP imaging, and gadolinium enhanced spin-echo T1-weighted imaging. Supplemental sagittal or coronal images were obtained, depending on lesion morphology and the patient’s clinical stability. We performed endoscopic examinations with this endoscope in 10 subjects, seven subjects with control and three subjects with gallbladder cancer. The study was accepted by the Ethics Committee of the institutions, and every participating patient gave an informed consent. Results: In the control subjects, we identified three tissue layers in the normal gallbladder. The mucosa and submucosa were depicted as the first layer with high intensity imaging, muscularis propria corresponded to the second layer with low intensity, and subserosa and serosa consisted the third layer with high intensity. On the basis of these findings by MR endoscopy, gallbladder tumors were identified and staged in all patients: subserosa invasion in all patients. A good histopathologic correlation was obtained. Conclusion: We are investigating the optimal combination of a phased-array surface coil with the endoscopic coil to improve signal-to-noise ratio in our images. We believe that there are important potential advantages with this technique in staging gallbladder cancer and in planning patient treatment.
A. Das, M. V. Sivak Jr., A. Chak, R. Ck Wong, V. Westphal, A. M. Rollins, J. Izatt, G. A. Isenberg, J. Willis, Univ Hospitals of Cleveland, Cleveland, OH; Case Western Reserve Univ, Cleveland, OH. Both OCT and CPEUS are candidates for high resolution imaging of the GI wall but their potential roles in clinical practice have not been investigated. It is unknown whether imaging data provided by the 2 modalities are complementary or duplicative. Aim: To determine the relative roles of OCT and CPEUS imaging. Methods: OCT and CPEUS were used to image normal appearing GI wall at same anatomic sites (random sequence) in patients undergoing elective procedures. Images were obtained using an OCT probe (2.4 mm in dia, 0.5 mm focal distance). CPEUS was done using a 20 MHz or a new 30 MHz catheter probe (UMS20 20R, XUM S30 25 R, Olympus) and water-filled balloon sheaths. Digitally captured images were rated for clarity, resolution, identification of layers of the wall and ability to identify microscopic structures. Results: To test depth of penetration and resolution in vitro, simultaneous OCT and CPEUS images were obtained in a phantom using a water submerged chamber and 3 monofilament fibers placed exactly at 1, 3 and 4 mm from the probes. CPEUS visualized all 3 fibers (estimated depth of penetration > 8 mm). OCT imaged the center and both sides of the nearest fiber (estimated penetration depth 1 mm). 44 histologically confirmed normal sites (10 esophagus, 9 stomach, 7 duodenum, 7 rectum, 11 colon) in 27 patients were evaluated. With OCT, mucosa and muscularis mucosa were clearly seen at all sites. Except for stomach, part of the submucosa was seen in all sites. OCT penetration ranged from from 0.7 to 0.9 mm; thickness of the mucosal layer varied from 0.4 mm (esophagus) to 0.5 mm (colon). Microscopic structures viz. esophageal glands, intestinal villi, colonic crypts and blood vessels were easily identified. CPEUS penetration ranged from 10 mm to 20 mm and 5 - 7 distinct layers were discerned. But, both mucosa and submucosa were seen as thin layers with no microscopic details. OCT was easier to perform because it does not require acoustic coupling. Conclusion: OCT resolution is superior to high frequency CPEUS, but depth of penetration is limited to mucosa and submucosa (mucosa only in stomach). OCT images the major structural components of the mucosa and submucosa whereas CPEUS does not. CPEUS images the full thickness of the gut wall, but the resolution is much less than OCT. Potentially, OCT and high frequency CPEUS could be complementary for clinical imaging.
*3434 ELECTROSTIMULATION TO MOVE ENDOSCOPES IN THE SMALL BOWEL. Mark N. Appleyard, Tim N. Mills, Sandy Mosse, Paul Swain, Gastrointestinal Sci Research Unit, London, United Kingdom; Univ Coll, London, United Kingdom; Royal London Hosp, London, United Kingdom. Background: Methods are required for tip propulsion of endoscopes passing through the small bowel and for propelling miniature capsule endoscopes without cables. Aim: To test the hypothesis that electrical stimulation could propel an endoscope by stimulating muscular contraction. Methods: Prototype devices of ovoid shape were constructed with electrodes mounted on the tapered section of the device. When in contact with the bowel wall electrical stimulation was applied causing circular muscle contraction of the adjacent muscle which when applied to the taper of the ovoid resulted in forward propulsion of the device. The device was connected by wires to an electrical generator allowing a range of currents and voltages to be tested. The device was tested in the small bowel (non-striated muscle) and in the oesophagus (transition of striated to non striated muscle) of the anaesthetised pig. Results: In initial tests in the pig oesophagus electrostimulation caused the ovoid to advance rapidly (6 mm/sec) down the oesophagus by inducing circular oesophageal muscle contraction. To our surprise the device moved at the same speed back up the oesophagus when reversed. Testing in the small bowel showed that this device was capable of moving in both directions and was capable of negotiating tight curves which could not be passed by conventional enteroscopes. Measured rates of travel at optimal settings were 3-4.5 mm/sec. The method does not induce peristalsis but works by stimulating local contraction. The current of 10 milliamps required for optimal movement seems to be below the threshold current that is reported as being perceived as painful in the human bowel although this remains to be demonstrated in man. The animals showed no ill effects and no effect on the cardiac cycle was noted during stimulation. A hearing aid type battery could supply power to keep this device moving in the small bowel for several hours. Conclusion: Electrostimulation might be used to move endoscopes in the small bowel.
*3436 ENHANCED MAGNIFICATION ENDOSCOPY: A NEW TECHNIQUE TO IDENTIFY SPECIALIZED INTESTINAL METAPLASIA IN BARRETT’S ESOPHAGUS. Moises Guelrud, Idamys Herrera, Julio Castro, Harold Essenfeld, Policlinica Metropolitana, Caracas, Venezuela. Specialized intestinal metaplasia (SIM) in Barrett´s esophagus (BE) is not identifiable by standard endoscopy (SE). Magnification endoscopy (ME) may be useful in the endoscopic recognition of BE. Acetic acid (AA) instillation improves visualization of columnar epithelium at the squamocolumnar union (Gastrointest Endosc 1998;47:512). Enhanced magnification endoscopy (EME) combines the use of ME and AA. Aim: To evaluate the usefulness of EME for the detection of SIM in BE. Methods: Patients undergoing endoscopic surveillance of BE underwent EME with an Olympus GIF-200Z (zoom 35X) and 1.5% AA instillation. SE was followed by ME and repeated after AA spraying. AA surface patterns were characterized prior to AA-directed biopsy of suspected areas and compare with histology results. A pathologist blinded to the biopsy methods read each biopsy. Logistic regression was performed to determine if surface patterns could predict SIM in suspected areas. Results: 49 patients with short segment BE were studied. One was excluded due to unclear definition of the surface pattern after EME. Four different endoscopic patterns were identified. I: round pits, II: reticular, III: villous, and IV: ridged. The table shows the percentage of endoscopic patterns observed with each endoscopic technique. The yield of detecting SIM according to the endoscopic pattern were: Pattern I:0%, II:11%(2/18), III:87%(40/46), and IV:100% (17/17). Odds ratio for Pattern IV for detecting SIM compared with Pattern I was 14 (p=0.015), ROC=0.95. Sensitivity, specificity, and positive and negative predicted values are 96.6%, 90.2%, 89.0%, and 97.0%. Total accuracy 93.1%. Conclusions: EME is an accurate method of predicting SIM in BE. It appears to be able to distinguish gastric epithelium from SIM and may be useful in directing biopsies and identifying SIM before and after ablation therapy.
*3435 HIGH RESOLUTION ENDOSCOPIC IMAGING OF THE GASTROINTESTINAL TRACT USING OPTICAL COHERENCE TOMOGRAPHY AND HIGH FREQUENCY CATHETER PROBE ENDOSCOPIC ULTRASONOGRAPHY .
VOLUME 51, NO. 4, PART 2, 2000
Patterns
Total areas#
SE
SE+AA
ME
EME
I II III IV
48 18 46 17
0 0 2% 6%
0 0 13% 29%
21% 39% 46% 65%
100% 100% 100% 100%
GASTROINTESTINAL ENDOSCOPY
AB91
A. Das, M. V. Sivak Jr., A. Chak, R. Ck Wong, V. Westphal, A. M. Rollins, J. Izatt, G. A. Isenberg, J. Willis, Univ Hospitals of Cleveland, Cleveland, OH; Case Western Reserve Univ, Cleveland, OH. Both OCT and CPEUS are candidates for high resolution imaging of the GI wall but their potential roles in clinical practice have not been investigated. It is unknown whether imaging data provided by the 2 modalities are complementary or duplicative. Aim: To determine the relative roles of OCT and CPEUS imaging. Methods: OCT and CPEUS were used to image normal appearing GI wall at same anatomic sites (random sequence) in patients undergoing elective procedures. Images were obtained using an OCT probe (2.4 mm in dia, 0.5 mm focal distance). CPEUS was done using a 20 MHz or a new 30 MHz catheter probe (UMS20 20R, XUM S30 25 R, Olympus) and water-filled balloon sheaths. Digitally captured images were rated for clarity, resolution, identification of layers of the wall and ability to identify microscopic structures. Results: To test depth of penetration and resolution in vitro, simultaneous OCT and CPEUS images were obtained in a phantom using a water submerged chamber and 3 monofilament fibers placed exactly at 1, 3 and 4 mm from the probes. CPEUS visualized all 3 fibers (estimated depth of penetration > 8 mm). OCT imaged the center and both sides of the nearest fiber (estimated penetration depth 1 mm). 44 histologically confirmed normal sites (10 esophagus, 9 stomach, 7 duodenum, 7 rectum, 11 colon) in 27 patients were evaluated. With OCT, mucosa and muscularis mucosa were clearly seen at all sites. Except for stomach, part of the submucosa was seen in all sites. OCT penetration ranged from from 0.7 to 0.9 mm; thickness of the mucosal layer varied from 0.4 mm (esophagus) to 0.5 mm (colon). Microscopic structures viz. esophageal glands, intestinal villi, colonic crypts and blood vessels were easily identified. CPEUS penetration ranged from 10 mm to 20 mm and 5 - 7 distinct layers were discerned. But, both mucosa and submucosa were seen as thin layers with no microscopic details. OCT was easier to perform because it does not require acoustic coupling. Conclusion: OCT resolution is superior to high frequency CPEUS, but depth of penetration is limited to mucosa and submucosa (mucosa only in stomach). OCT images the major structural components of the mucosa and submucosa whereas CPEUS does not. CPEUS images the full thickness of the gut wall, but the resolution is much less than OCT. Potentially, OCT and high frequency CPEUS could be complementary for clinical imaging.
*3434 ELECTROSTIMULATION TO MOVE ENDOSCOPES IN THE SMALL BOWEL. Mark N. Appleyard, Tim N. Mills, Sandy Mosse, Paul Swain, Gastrointestinal Sci Research Unit, London, United Kingdom; Univ Coll, London, United Kingdom; Royal London Hosp, London, United Kingdom. Background: Methods are required for tip propulsion of endoscopes passing through the small bowel and for propelling miniature capsule endoscopes without cables. Aim: To test the hypothesis that electrical stimulation could propel an endoscope by stimulating muscular contraction. Methods: Prototype devices of ovoid shape were constructed with electrodes mounted on the tapered section of the device. When in contact with the bowel wall electrical stimulation was applied causing circular muscle contraction of the adjacent muscle which when applied to the taper of the ovoid resulted in forward propulsion of the device. The device was connected by wires to an electrical generator allowing a range of currents and voltages to be tested. The device was tested in the small bowel (non-striated muscle) and in the oesophagus (transition of striated to non striated muscle) of the anaesthetised pig. Results: In initial tests in the pig oesophagus electrostimulation caused the ovoid to advance rapidly (6 mm/sec) down the oesophagus by inducing circular oesophageal muscle contraction. To our surprise the device moved at the same speed back up the oesophagus when reversed. Testing in the small bowel showed that this device was capable of moving in both directions and was capable of negotiating tight curves which could not be passed by conventional enteroscopes. Measured rates of travel at optimal settings were 3-4.5 mm/sec. The method does not induce peristalsis but works by stimulating local contraction. The current of 10 milliamps required for optimal movement seems to be below the threshold current that is reported as being perceived as painful in the human bowel although this remains to be demonstrated in man. The animals showed no ill effects and no effect on the cardiac cycle was noted during stimulation. A hearing aid type battery could supply power to keep this device moving in the small bowel for several hours. Conclusion: Electrostimulation might be used to move endoscopes in the small bowel.
*3436 ENHANCED MAGNIFICATION ENDOSCOPY: A NEW TECHNIQUE TO IDENTIFY SPECIALIZED INTESTINAL METAPLASIA IN BARRETT’S ESOPHAGUS. Moises Guelrud, Idamys Herrera, Julio Castro, Harold Essenfeld, Policlinica Metropolitana, Caracas, Venezuela. Specialized intestinal metaplasia (SIM) in Barrett´s esophagus (BE) is not identifiable by standard endoscopy (SE). Magnification endoscopy (ME) may be useful in the endoscopic recognition of BE. Acetic acid (AA) instillation improves visualization of columnar epithelium at the squamocolumnar union (Gastrointest Endosc 1998;47:512). Enhanced magnification endoscopy (EME) combines the use of ME and AA. Aim: To evaluate the usefulness of EME for the detection of SIM in BE. Methods: Patients undergoing endoscopic surveillance of BE underwent EME with an Olympus GIF-200Z (zoom 35X) and 1.5% AA instillation. SE was followed by ME and repeated after AA spraying. AA surface patterns were characterized prior to AA-directed biopsy of suspected areas and compare with histology results. A pathologist blinded to the biopsy methods read each biopsy. Logistic regression was performed to determine if surface patterns could predict SIM in suspected areas. Results: 49 patients with short segment BE were studied. One was excluded due to unclear definition of the surface pattern after EME. Four different endoscopic patterns were identified. I: round pits, II: reticular, III: villous, and IV: ridged. The table shows the percentage of endoscopic patterns observed with each endoscopic technique. The yield of detecting SIM according to the endoscopic pattern were: Pattern I:0%, II:11%(2/18), III:87%(40/46), and IV:100% (17/17). Odds ratio for Pattern IV for detecting SIM compared with Pattern I was 14 (p=0.015), ROC=0.95. Sensitivity, specificity, and positive and negative predicted values are 96.6%, 90.2%, 89.0%, and 97.0%. Total accuracy 93.1%. Conclusions: EME is an accurate method of predicting SIM in BE. It appears to be able to distinguish gastric epithelium from SIM and may be useful in directing biopsies and identifying SIM before and after ablation therapy.
*3435 HIGH RESOLUTION ENDOSCOPIC IMAGING OF THE GASTROINTESTINAL TRACT USING OPTICAL COHERENCE TOMOGRAPHY AND HIGH FREQUENCY CATHETER PROBE ENDOSCOPIC ULTRASONOGRAPHY .
VOLUME 51, NO. 4, PART 2, 2000
Patterns
Total areas#
SE
SE+AA
ME
EME
I II III IV
48 18 46 17
0 0 2% 6%
0 0 13% 29%
21% 39% 46% 65%
100% 100% 100% 100%
GASTROINTESTINAL ENDOSCOPY
AB91