- Email: [email protected]
Endoscopic ultrasound (EUS) evaluation of perianal fistulas: Comparing two probes and assessing the role of hydrogen peroxide injection
T1736
CC. Results 9 patients with crohns disease (CD) and 7 patients with ukerative colitis (UC) were evaluated Overall 101 of 112 colon parts (902%) were filled and depicted adequately tar diagnosia. In patients with CD there were no pathological flndiugs in 2 patients with both methods. In 7 patients inflammation of the caecum or typieall skip lesions were assessed with MRC and CC In one of these patients inflammation of the sigma and a stenosis, which could be passed endoscopically, were assessed by CC, bnt not with MR(:. In patients with CD MR(: had a sensitivity of 89% with a specifity of 100% tar all diagnosed lesions using CC as gnld standard~ In 3 paoems with UC mild inflammation of the colon was diagnosed endoscopically, which was not seen in MR(:. One patient with UC had an acute severe inflammation of the whole colon, not seen based on MRC. In 2 patients with intlanmlation of the whole cokin a f01d rarefication was noticed. Only in one patient endoscopicaily diagnosed mflarnmation of the sigma was seen as a ,,vail thickening wdth MRC. MR1 based diagnosis had a sensiti~qty of only 43% with a specihty of 100% Virtual endoscopy did not add aW relevant intonnation Raretication of bowel tolds was depicted impressively. Additionally pseudopolyps in the caecum in 2 patients were diagmosed based on VE, which were overlooked on 2D image readings. ( onclusion MRI based colonography appears to be usefi.d m the assessrnent and tallow up of patients with CD whereas in UC only a poor sensitivity can be achieved because at the mucosal character of disease
Prospective Comparison of Hydrogen Peroxide-Enhanced 3D endoanal Uhrasonography and Endoanal MR Imaging for Perianal Fistulas Rachel L West, Soender Dwarkasing, David D. E. Zimmerman, Shahid M. Hnasain, Wim C. J, Hop, Willem R. Schotuen, Ernst J. Kuipers, BachelleJ. F. Feh-Bersma Objectives: Successful fistula surgery depends on accurate preoperative assessment This is important so tit_at recurrence of fistulas does not occur and the continence mechanism is preserved. Magnetic resonance imaging (MR1) and anal endosonography (AES) ate both used to assess perianal fistulas and have an excellent correlation with surgical findings. Several studies have compared the results of MRI and AES. The resuhs in these studies differ However there are no studies comparing hydrogen peroxide-enhanced 3D endoanal uhrasonography (HPUS) and endoanal MR] The purpose of our study was to assess agreement between 3D HPUS and endoanal MRI in perlanal fistulas. Methods: Patients referred to the surgeon for perianal fistulas underwent 3D HPUS and endoanal MR1. The 3D HPUS and endoanal MRI results were assessed separately by experienced observers. Both were blinded tO1each other's findings. A description of each fistula was made and the following characteristins were recorded: classification according to Parks, presence of secondary tracts (circular or linear) and identification of toe internal opening. Results: in total 24 patients (20 M, aged 27-71) were included and underwent 3D HPUS and eudoanal MR1. The median time betsveen 3D HPUS and endoanal MRI was 66 days (IQR 21-160). The methods agreed in 92% (22/24) for the Parks classification, in 71% (17/24) for the circular secondary tracts, in 96% (23/24) for the linear secondary tracts, and in 92% (22/24) lot the internal opening. Conclusions: For evaluating periamal fistulas 3D HPUS and endoanal MRI have a good agreement, especially for the Parks classification and the internal opening. There is a poor agreement in evaluating the circular secondary tracts. Differences could be explained by the nine interval between 3D HPUS and endoanal MR1. 3D HPUS and endoanal MRI oftor comparable results in evaluating single perianal tlatulas and both can be used for reliable preoperative assessment.
T1734 Usefulness of MRI in Assessing The Necessity of Surgery for Acute Small-Bowel Obstruction gel Aoki~ "faro Takafiara, Yasuharu Yamaguchi, Yasuyuki [mao, Yasushi Morita, Miki Minra, Naoya Katsumi, Kenli Nakamura, Taro Yamam, Shin-Idli Takahashi [Objective] In acute small-bowel obstruction, it is very important to assess whether emergency surgery is required or a conservative follow-up is selected. However, the conventional diagnostic imaging has not always provided a satisfactory assessment. This time we have tested the usefultmss ok Clue-MR1 in assessnig the necessity of emergency surgeD'. [Sub[ectsl Thirty-eight cases (18 males and 20 t~emales)with an average age of 57.5 years were admitted to our hospital because of acute small-bowel obstruction tx:tweenJanuary, 2001 and October, 2002, and were examuled by MR1 at the onset of the disease. [Methods] The Philips or Toshiba I.ST MRI machine was used, and for Cine-MRI, Single-Shot High-Speed SE imaging methods or Balanced TEE were ernployed. Emergency surgery was required when ClueMR1 showed no penstalsis in the intestinal tract associated with obstruction such as dosed loop while showed peristalsis in the surrounding intestine (peristalsis gap sign; PG), and either of intralumniat bleednig sign (IB) exfiit)itnig low MRI signal on both the T2 weighted image and tim tht-suppressed 1"2 weighted image was positive. On the other hand, We choc~sed either palliative surgery or conservative treatment based on the MRI fniding that obstruction sites ,.,,'eremultiple and that a large amount of ascites was present in the cases assessed as requiring no emergency stugeD,. [Results] MR1 findings revealed that 12 cases required emeLgcncy surgery, 9 ca~s of which received surgery, with a 90% specificity and a 100% sensitivity In contrast, MR[ findings revealed that 26 cases required no emergency surgery, all cases of which actually eliminated the necessity of emergency surgery (NPV 100%). Thirteen cases expected to receive palliative surgery, 5 cases of which recewed palliative sin\gory. The auother thirteen cases were considered to conservatively improve, ordy 1 case of which received palliative surgery', and the remaining 12 cases of which impruved conservatively. [Conclusion] MRI findings allow to evaluate the necessity of eme> gency surgery with high specificity and setksitiviD. ni acute small-bowel obstruction. It is also considered that MN findings may be an exceIlent index of the prediction for conservative improvement
T1737 Endoscopic Ultrasound (EUS) Evaluation of Perianal Fistulas: Comparing Two Probes and Assessing the Role of Hydrogen Peroxide Injection Yaakov Maor, Yehuda Chowers, Benjamin Avidan, Moshe Koller, Oded Zmora, Simon Ba>Meir Introduction: EUS is one of the most accurate methods for the assessment of perianal fistulas. Conventional radial instruments are often used, Hydrogen peroxide (H202) injection into the fistula was claimed to improve fistula visualization. Aims: To compare conventional radial and dedicated rectal EUS probes for the evaluation of perianal fistulas and to determine the role of H202 injection in outlniing fistula tracts. Methods: Patients with perianal hstulas were evaluated using the GF-UM20 radial instrument (Olympus) and the EUP-R53 MonaPlane rectal probe (Bentax). External openings were located and assessed for active drainage. Fistula tracts were delineated according to Parks' classification. Abscesses were located and measured. Repeated evaluation was preformed following H202 ir~jection. Results: Thirtynine patients: male-28, female-11, age 40 (16-58) years, underwent 42 studies. Seventeen patients had 1BD: Crohn's disease-15, ulcerative colitis-2. Injection of H202 was the only way to visualize fistulas in 3 patients using EUP-R53; it miproved ~:sualization of hstula in 7 and 1 using EUP-R53 and GF-UM20, respectively. Injection of H202 improved visualization of abscesses m 3 and 1 using EUP-R53 and GF-UM20, respectively, Conclusions: EUPR53 rectal probe is superior to the GF-UM20 instrument in the evaluation of perianal fistulas. H202 injection improved fistula visualization and probably the therapeutic approach
T1735
GF,UM20 and EUP.R53for detectionof fistulas and abscesses Rstulas Detected
Radiologic Assessment of the Upper GI Tract: Does it Play a Role in Determining Bariatric Surgical Approach? Ravi N Sharaf Elizabeth H~ Weinshel, Edmund J. Bnii, Jonathan Rosenberg, Christine Pen
TransSphlncte~
~ Sphinct~c
De~Aed
Su~aSupecRclal Levator
EUPR53
27
1
19
7
19
15
4
GF.
7
0
6
1
2
2
0
UM~
BACKGROUND:The role of upper GI series (UGIS) prior to bariatric surgery' is controversial. l'he aim of tbis study was to evaluate the diagnostic yield and cost effectiveness of routine UGIS prior to bariatric surgery. METHODS: The medical records of consecutive obese patients who nnderwent UGIS prior to bariatric surgery between 4/01 and 10/02 were reviewed UGIS reports were reviewed by two experienced gastroenterologists and the findings were divided into 4 groups based on predetermined criteria: group 0 (normal stud),'), group 1 (abnormal findings that did not change surgical approach), group 2 (abnormal findings that did change surgical approach m"postpone surgery), and goup 3 (results which were an absolute contraindication to surgery), Clinically important findings included lesions in groups 2 and 3. The cost of an upper G1 series ($10080) was estimated from the physician tee under Medicaid reimbursement. RESULTS: During the 18-month study' period, 171 patients were evaluated by UGIS prior to bariau'ic smgmT The mean age of the patients was 39.4 +/- 10.6 years, 667% were ti'male, 825% were Caucasian, and the mean BMI was 482 +/- 78 kg/m2 Upper GI symptoms were" present in 29.2%, and 14.0% were taking a H2 blocker or proton pump irihibitor. One or more lesions were identified in 48.0% of patients, ~;~ith only 5.3% having clinically important findings Tire prevalence of ra&ologic findings using the classification system above was as tallows: group 0 (52.0%), group I (42 7%), group 2 (5 3%), and group 3 (0.0%). The most common findings identified were esopMgeal reflux (21 6%), hiataI hernias (18.7%), Schatzkl's ring (1,8%), and esophagitis (1 2%). There were no significant diffi:rences between patients with and those without clinically important findings with regard to age, gender, race, BMI, G1 symptoms, H2 or proton pump inhibitor use, smoking, alcohol use, or hemoglobin levels. The cost of performing rmltine UGIS on all patients prior to bariatric surgery was $1,915.20 per clinically important finding detected, The type of bariatric surgery perfomled did not differ between patients with and those without clinically important findings (p = 0.78) although the ma/onty at the patients had laparoscopic adjustable gastric banding (93%), CONCLUSIONS: In this retrospective study, routine UG1S prior to bariatric surgery had a kiw diagnostic ymld, however, it changed the surgical approach o1"delayed surgeD*in 5% of patients. "*Veildesigned prospective studies to address these issues are warranted.
Abscesm~ Sapedlcial
T1738 Role of Endoscopic Ultrasound Guided Fine Needle Aspiration in the Diagnosis of Extrahepatic Biliary Tract and Gallbladder Lesions Jan R Steinhauer, Darshana N Jhala, Mahmoud A. El Oubeidi, lsam A. Eltoum, David C Chhieng, Ann P. Loam, Nirag C Jhala Intruduction: Ultrasound guided, percutaneous fine needle aspiration (FNA) is infrequently used tor the evaluation of mass lesions of the biliary tract. Endoscopic ultrasound (EUS) guicted FNA is used more frequently kir the diagnosis of pancreaticobihary tract malignancies with a high level of diagnostic accuracy. However, its utility in the evaluation of lesions of the gallbladder and extrahepatic bile duct has not been ,,yell investigated. At our institution, EUS-FNA is combined with on-site interpretation to provide an immediate feedback on specimen adequacy and a provisional diagnosis. We report our experience of EUS-FNA in the diagnosis of lesions of the gallbladder and the extrahepatic bile duct at our institution. Materials and Methods: All EUS-FNA (n = 29) specimei~sof the extrahepatic bile duct (n = 25) and gallbladder (n = 4) fl'om 27 patients (17 males aged 37-81 years and 10 females aged 49-74 }'ears) over a 22 month penod were evaluated. All reports were broadly categnrized as positive or negative for malignancy, atypical or unsadsfactory. Tissue diagnosis and clinical Mlow-up information were correlated with cytological resnhs, and frequency of procedural complications was recorded. Sensitivity, specificity and diagnostic accuracy of each method were detem'lined. Results: Eighteen (62%), 8 (28%) and 3 (10%) EUS-FNA samples were characterized as positive for malignancy, negatwe {or malignancy and atypical, respectively. No unsatisfactory"speciulens ,,,,'ereidentified (0%). Of the three atypical results, two patients
A-557
AGA Abstracts
CC. Results 9 patients with crohns disease (CD) and 7 patients with ukerative colitis (UC) were evaluated Overall 101 of 112 colon parts (902%) were filled and depicted adequately tar diagnosia. In patients with CD there were no pathological flndiugs in 2 patients with both methods. In 7 patients inflammation of the caecum or typieall skip lesions were assessed with MRC and CC In one of these patients inflammation of the sigma and a stenosis, which could be passed endoscopically, were assessed by CC, bnt not with MR(:. In patients with CD MR(: had a sensitivity of 89% with a specifity of 100% tar all diagnosed lesions using CC as gnld standard~ In 3 paoems with UC mild inflammation of the colon was diagnosed endoscopically, which was not seen in MR(:. One patient with UC had an acute severe inflammation of the whole colon, not seen based on MRC. In 2 patients with intlanmlation of the whole cokin a f01d rarefication was noticed. Only in one patient endoscopicaily diagnosed mflarnmation of the sigma was seen as a ,,vail thickening wdth MRC. MR1 based diagnosis had a sensiti~qty of only 43% with a specihty of 100% Virtual endoscopy did not add aW relevant intonnation Raretication of bowel tolds was depicted impressively. Additionally pseudopolyps in the caecum in 2 patients were diagmosed based on VE, which were overlooked on 2D image readings. ( onclusion MRI based colonography appears to be usefi.d m the assessrnent and tallow up of patients with CD whereas in UC only a poor sensitivity can be achieved because at the mucosal character of disease
Prospective Comparison of Hydrogen Peroxide-Enhanced 3D endoanal Uhrasonography and Endoanal MR Imaging for Perianal Fistulas Rachel L West, Soender Dwarkasing, David D. E. Zimmerman, Shahid M. Hnasain, Wim C. J, Hop, Willem R. Schotuen, Ernst J. Kuipers, BachelleJ. F. Feh-Bersma Objectives: Successful fistula surgery depends on accurate preoperative assessment This is important so tit_at recurrence of fistulas does not occur and the continence mechanism is preserved. Magnetic resonance imaging (MR1) and anal endosonography (AES) ate both used to assess perianal fistulas and have an excellent correlation with surgical findings. Several studies have compared the results of MRI and AES. The resuhs in these studies differ However there are no studies comparing hydrogen peroxide-enhanced 3D endoanal uhrasonography (HPUS) and endoanal MR] The purpose of our study was to assess agreement between 3D HPUS and endoanal MRI in perlanal fistulas. Methods: Patients referred to the surgeon for perianal fistulas underwent 3D HPUS and endoanal MR1. The 3D HPUS and endoanal MRI results were assessed separately by experienced observers. Both were blinded tO1each other's findings. A description of each fistula was made and the following characteristins were recorded: classification according to Parks, presence of secondary tracts (circular or linear) and identification of toe internal opening. Results: in total 24 patients (20 M, aged 27-71) were included and underwent 3D HPUS and eudoanal MR1. The median time betsveen 3D HPUS and endoanal MRI was 66 days (IQR 21-160). The methods agreed in 92% (22/24) for the Parks classification, in 71% (17/24) for the circular secondary tracts, in 96% (23/24) for the linear secondary tracts, and in 92% (22/24) lot the internal opening. Conclusions: For evaluating periamal fistulas 3D HPUS and endoanal MRI have a good agreement, especially for the Parks classification and the internal opening. There is a poor agreement in evaluating the circular secondary tracts. Differences could be explained by the nine interval between 3D HPUS and endoanal MR1. 3D HPUS and endoanal MRI oftor comparable results in evaluating single perianal tlatulas and both can be used for reliable preoperative assessment.
T1734 Usefulness of MRI in Assessing The Necessity of Surgery for Acute Small-Bowel Obstruction gel Aoki~ "faro Takafiara, Yasuharu Yamaguchi, Yasuyuki [mao, Yasushi Morita, Miki Minra, Naoya Katsumi, Kenli Nakamura, Taro Yamam, Shin-Idli Takahashi [Objective] In acute small-bowel obstruction, it is very important to assess whether emergency surgery is required or a conservative follow-up is selected. However, the conventional diagnostic imaging has not always provided a satisfactory assessment. This time we have tested the usefultmss ok Clue-MR1 in assessnig the necessity of emergency surgeD'. [Sub[ectsl Thirty-eight cases (18 males and 20 t~emales)with an average age of 57.5 years were admitted to our hospital because of acute small-bowel obstruction tx:tweenJanuary, 2001 and October, 2002, and were examuled by MR1 at the onset of the disease. [Methods] The Philips or Toshiba I.ST MRI machine was used, and for Cine-MRI, Single-Shot High-Speed SE imaging methods or Balanced TEE were ernployed. Emergency surgery was required when ClueMR1 showed no penstalsis in the intestinal tract associated with obstruction such as dosed loop while showed peristalsis in the surrounding intestine (peristalsis gap sign; PG), and either of intralumniat bleednig sign (IB) exfiit)itnig low MRI signal on both the T2 weighted image and tim tht-suppressed 1"2 weighted image was positive. On the other hand, We choc~sed either palliative surgery or conservative treatment based on the MRI fniding that obstruction sites ,.,,'eremultiple and that a large amount of ascites was present in the cases assessed as requiring no emergency stugeD,. [Results] MR1 findings revealed that 12 cases required emeLgcncy surgery, 9 ca~s of which received surgery, with a 90% specificity and a 100% sensitivity In contrast, MR[ findings revealed that 26 cases required no emergency surgery, all cases of which actually eliminated the necessity of emergency surgery (NPV 100%). Thirteen cases expected to receive palliative surgery, 5 cases of which recewed palliative sin\gory. The auother thirteen cases were considered to conservatively improve, ordy 1 case of which received palliative surgery', and the remaining 12 cases of which impruved conservatively. [Conclusion] MRI findings allow to evaluate the necessity of eme> gency surgery with high specificity and setksitiviD. ni acute small-bowel obstruction. It is also considered that MN findings may be an exceIlent index of the prediction for conservative improvement
T1737 Endoscopic Ultrasound (EUS) Evaluation of Perianal Fistulas: Comparing Two Probes and Assessing the Role of Hydrogen Peroxide Injection Yaakov Maor, Yehuda Chowers, Benjamin Avidan, Moshe Koller, Oded Zmora, Simon Ba>Meir Introduction: EUS is one of the most accurate methods for the assessment of perianal fistulas. Conventional radial instruments are often used, Hydrogen peroxide (H202) injection into the fistula was claimed to improve fistula visualization. Aims: To compare conventional radial and dedicated rectal EUS probes for the evaluation of perianal fistulas and to determine the role of H202 injection in outlniing fistula tracts. Methods: Patients with perianal hstulas were evaluated using the GF-UM20 radial instrument (Olympus) and the EUP-R53 MonaPlane rectal probe (Bentax). External openings were located and assessed for active drainage. Fistula tracts were delineated according to Parks' classification. Abscesses were located and measured. Repeated evaluation was preformed following H202 ir~jection. Results: Thirtynine patients: male-28, female-11, age 40 (16-58) years, underwent 42 studies. Seventeen patients had 1BD: Crohn's disease-15, ulcerative colitis-2. Injection of H202 was the only way to visualize fistulas in 3 patients using EUP-R53; it miproved ~:sualization of hstula in 7 and 1 using EUP-R53 and GF-UM20, respectively. Injection of H202 improved visualization of abscesses m 3 and 1 using EUP-R53 and GF-UM20, respectively, Conclusions: EUPR53 rectal probe is superior to the GF-UM20 instrument in the evaluation of perianal fistulas. H202 injection improved fistula visualization and probably the therapeutic approach
T1735
GF,UM20 and EUP.R53for detectionof fistulas and abscesses Rstulas Detected
Radiologic Assessment of the Upper GI Tract: Does it Play a Role in Determining Bariatric Surgical Approach? Ravi N Sharaf Elizabeth H~ Weinshel, Edmund J. Bnii, Jonathan Rosenberg, Christine Pen
TransSphlncte~
~ Sphinct~c
De~Aed
Su~aSupecRclal Levator
EUPR53
27
1
19
7
19
15
4
GF.
7
0
6
1
2
2
0
UM~
BACKGROUND:The role of upper GI series (UGIS) prior to bariatric surgery' is controversial. l'he aim of tbis study was to evaluate the diagnostic yield and cost effectiveness of routine UGIS prior to bariatric surgery. METHODS: The medical records of consecutive obese patients who nnderwent UGIS prior to bariatric surgery between 4/01 and 10/02 were reviewed UGIS reports were reviewed by two experienced gastroenterologists and the findings were divided into 4 groups based on predetermined criteria: group 0 (normal stud),'), group 1 (abnormal findings that did not change surgical approach), group 2 (abnormal findings that did change surgical approach m"postpone surgery), and goup 3 (results which were an absolute contraindication to surgery), Clinically important findings included lesions in groups 2 and 3. The cost of an upper G1 series ($10080) was estimated from the physician tee under Medicaid reimbursement. RESULTS: During the 18-month study' period, 171 patients were evaluated by UGIS prior to bariau'ic smgmT The mean age of the patients was 39.4 +/- 10.6 years, 667% were ti'male, 825% were Caucasian, and the mean BMI was 482 +/- 78 kg/m2 Upper GI symptoms were" present in 29.2%, and 14.0% were taking a H2 blocker or proton pump irihibitor. One or more lesions were identified in 48.0% of patients, ~;~ith only 5.3% having clinically important findings Tire prevalence of ra&ologic findings using the classification system above was as tallows: group 0 (52.0%), group I (42 7%), group 2 (5 3%), and group 3 (0.0%). The most common findings identified were esopMgeal reflux (21 6%), hiataI hernias (18.7%), Schatzkl's ring (1,8%), and esophagitis (1 2%). There were no significant diffi:rences between patients with and those without clinically important findings with regard to age, gender, race, BMI, G1 symptoms, H2 or proton pump inhibitor use, smoking, alcohol use, or hemoglobin levels. The cost of performing rmltine UGIS on all patients prior to bariatric surgery was $1,915.20 per clinically important finding detected, The type of bariatric surgery perfomled did not differ between patients with and those without clinically important findings (p = 0.78) although the ma/onty at the patients had laparoscopic adjustable gastric banding (93%), CONCLUSIONS: In this retrospective study, routine UG1S prior to bariatric surgery had a kiw diagnostic ymld, however, it changed the surgical approach o1"delayed surgeD*in 5% of patients. "*Veildesigned prospective studies to address these issues are warranted.
Abscesm~ Sapedlcial
T1738 Role of Endoscopic Ultrasound Guided Fine Needle Aspiration in the Diagnosis of Extrahepatic Biliary Tract and Gallbladder Lesions Jan R Steinhauer, Darshana N Jhala, Mahmoud A. El Oubeidi, lsam A. Eltoum, David C Chhieng, Ann P. Loam, Nirag C Jhala Intruduction: Ultrasound guided, percutaneous fine needle aspiration (FNA) is infrequently used tor the evaluation of mass lesions of the biliary tract. Endoscopic ultrasound (EUS) guicted FNA is used more frequently kir the diagnosis of pancreaticobihary tract malignancies with a high level of diagnostic accuracy. However, its utility in the evaluation of lesions of the gallbladder and extrahepatic bile duct has not been ,,yell investigated. At our institution, EUS-FNA is combined with on-site interpretation to provide an immediate feedback on specimen adequacy and a provisional diagnosis. We report our experience of EUS-FNA in the diagnosis of lesions of the gallbladder and the extrahepatic bile duct at our institution. Materials and Methods: All EUS-FNA (n = 29) specimei~sof the extrahepatic bile duct (n = 25) and gallbladder (n = 4) fl'om 27 patients (17 males aged 37-81 years and 10 females aged 49-74 }'ears) over a 22 month penod were evaluated. All reports were broadly categnrized as positive or negative for malignancy, atypical or unsadsfactory. Tissue diagnosis and clinical Mlow-up information were correlated with cytological resnhs, and frequency of procedural complications was recorded. Sensitivity, specificity and diagnostic accuracy of each method were detem'lined. Results: Eighteen (62%), 8 (28%) and 3 (10%) EUS-FNA samples were characterized as positive for malignancy, negatwe {or malignancy and atypical, respectively. No unsatisfactory"speciulens ,,,,'ereidentified (0%). Of the three atypical results, two patients
A-557
AGA Abstracts