- Email: [email protected]
Prospective Trial of Endoscopic Clips versus Combination Therapy in Upper GI Bleeding (PROTECCT—UGI Bleeding)
**602 Utility of EUS-FNA in the Evaluation of Primary Lung Mass Shyam Varadarajulu, Brenda J. Hoffman, Robert H. Hawes, Mohamad Eloubeidi Introduction: The accuracy, safety, and cost-effectiveness of EUS-FNA in screening patients with lung cancer for mediastinal metastasis are well established. While few studies have reported that FNA of mediastinal mass in some cases yield a diagnosis of lung cancer, the utility of EUS-FNA in evaluating primary lung mass per se has not been reported. Aim: To determine the diagnostic yield and safety of EUS-FNA in evaluating primary lung mass lesions. Methods: A database search was conducted for all patients who underwent EUS-FNA of a primary lung mass. The indications for procedure, technical outcomes, complications, and clinical follow-up were obtained by review of patient records. Results: Eighteen patients (11 male, 7 female) underwent EUS-FNA of a primary lung mass over a 3yr period (2000-2003).The indication for EUS was mediastinal staging of suspected lung cancer noted on chest-CT. Fourteen patients (78%) had right or left side lower lobe lung mass and 4 (22%) had right or left side upper lobe lung mass. All patients had failed prior attempts by CT-guided biopsy and/or bronchoscopy to establish tissue diagnosis of the primary lung mass. FNA of a lung mass was done when no mediastinal lymph node was available for sampling or if after three passes on a mediastinal lymph node no preliminary diagnosis was established or if the lymph node stations could not be well visualized due to mediastinal invasion. EUS-FNA yielded tissue for diagnosis in 100% of patients: 15 had non-small cell lung cancer, 1 small cell lung cancer, and 2 had metastatic cancer. Ten of 18 patients had mediastinal invasion (T4) and in eight the mass was confined to the lung parenchyma. The mean number of passes to establish a diagnosis was 2 (range, 1-6). No complications were encountered at a mean follow-up of 141 days (range, 72-396). Five patients underwent curative surgery and 13 were palliated with chemoradiation. Conclusions: EUS-FNA of primary lung mass is safe and establishes a diagnosis in 100% of patients with lesions that are accessible for fine needle aspiration. This may be particularly relevant for patients with suspected lung cancer but without mediastinal adenopathy.
****612 Prospective Trial of Endoscopic Clips versus Combination Therapy in Upper GI Bleeding (PROTECCT—UGI Bleeding) John Saltzman, Shelley Walsh, Veruska Di Sena, Christopher Huang, Benjamin Merrifield, Lisa Strate, Rie Ookubo, David Carr-Locke Purpose: Combination therapy (CT) with injection of epinephrine and multi-polar electrocoagulation is a common endoscopic treatment for non-variceal UGI bleeding. Endoscopic hemoclips (HC) have theoretical advantages over CT and are also widely used as a primary treatment. The aim of this study is to compare the efficacy of HC versus CT in the treatment of acute non-variceal UGI bleeding. Methods: Adult outpatients and inpatients with UGI bleeding at a large university hospital were eligible to participate. Inclusion criteria were active non-variceal UGI bleeding, a non-bleeding visible vessel (NBVV), or active bleeding or a NBVV after the mechanical removal of an adherent clot. Patients were randomized to either HC or CT at the time of the decision to perform endoscopic therapy. All outcomes were assessed including rebleeding and death 30 days after initial therapy. Results: Forty-seven patients were studied: 26 patients randomized to HC and 21 to CT. The mean age, sex, ethnicity, comorbidity scores (Charleston Index), and use of NSAID or anticoagulant drugs were not different between the two groups. The most common endoscopic finding in both groups was an ulcer (81% HC and 86% CT patients). There were 22 patients with active bleeding (14 HC, 8 CT, p=0.55), 13 with a NBVV (6 HC, 7 CT), and 12 with an adherent clot (6 HC, 6 CT). Primary hemostasis was similar between the two groups with successful initial control of bleeding in 26/26 HC patients (100%) and 20/21 (95.2%) CT patients, p=0.45. The rebleeding rates were also similar in both groups: 4/26 HC patients (15.4%) versus 5/21 CT patients (23.8%), p=0.49. In the patients who rebled, no association was demonstrated between stigmata of hemorrhage and treatment arm, p=0.19. Overall for HC vs. CT patients the median length of hospital stay (4 6 2 vs. 4 6 0.8 days), median units of blood transfused (4 6 1 vs. 5 6 1 units), surgery rates (3 vs. 1 patient), and mortality (0 vs. 2 patients, none directly from GI bleeding) were not different. The median duration of endoscopic therapy was 20 6 2 minutes in the HC group vs. 20 6 5 minutes in the CT group, p=0.45. One patient (CT group) developed aspiration pneumonia as a complication of the endoscopic procedure. Conclusions: In this first prospective, randomized controlled trial of endoscopic hemoclips versus combination therapy in nonvariceal upper GI bleeding, the efficacy and complications to date of the two treatment modalities are similar.
VOLUME 59, NO. 5, 2004
****613 Endoscopic Variceal Ligation (EVL): The More, the Merrier? A Prospective Multicenter Randomized Controlled Study Jose D. Sollano Jr., Frederick T. Dy, Stephen N. Wong, Melchor M. Chan, Albert E. Ismael, Jose A. Tan, Evan G. Ong, Marvin D. Basco Background: Bleeding from ruptured esophageal varices is a catastrophic sequela of portal hypertension and is associated with a high mortality rate. Endoscopic variceal ligation (EVL) can completely obliterate varices and is currently the treatment of choice for acute variceal bleeding. Due to its low complication rate and ease of deployment, ligating as many varices as possible in a single session may improve outcomes and cut costs of therapy. The objective of this study is to determine the outcome of deploying more than 6 bands vs. 6 bands or less per session of EVL. Methods: All patients with evidence of active or recent bleeding from esophageal varices with no history of previous EVL, sclerotherapy or intake of beta-blockers for the past 6 months were randomized into two groups (Group 1 >6 bands; Group 2 <6 bands; Six Shooter Saeed Multi-band Ligator, WilsonCook, Winston-Salem, NC.). Repeat endoscopies were done at day 10 and at 7-day intervals thereafter until all varices were obliterated. Patients did not receive additional medications. Follow-up endoscopies were scheduled every 6 months. Primary endpoints were complete variceal obliteration and rebleeding. Mortality was considered as a secondary endpoint. Intent-to-treat analysis was done for all endpoints. Results: A total of 59 patients were randomized (Group 1=28; Group 2=31). Baseline characteristics were similar (p>0.05) although there was more alcoholic cirrhosis in Group 2 (Group 1=21.4% vs. Group 2=51.6%, p=0.03). Mean follow-up was 13.5 +/ÿ 9.7 mos. for all patients. Sessions needed for complete variceal obliteration were similar (Group 1=1.6 vs. Group 2=1.9, p=0.209). There was a failure to achieve complete obliteration in 18% and 13% of patients in Groups 1 and 2, respectively. There was no difference in both rebleeding (42.9% vs. 25.8% overall rebleeding; mean interval to rebleeding 13.6 vs. 20 mos.; 2-year rebleeding rate 65.3% vs. 31.8%; Group 1 vs. 2, p=0.182) and survival between the 2 groups (60.7% vs. 71% overall survival; 17.4 vs. 19.7 mos. mean survival; 40% vs. 69% 2-year survival rate; Group 1 vs. 2, p=0.439). On multivariate analysis, female sex was associated with higher rebleeding rates while Child-Pugh class, rebleeding and failure to obliterate the varices were associated with higher mortality rates. Conclusions: In EVL, deploying more than 6 bands per session does not reduce the sessions needed for variceal obliteration and offers no added advantage in reducing rebleeding and mortality rates.
**673 Women's Preference for Women Physicians Is a Barrier to Colorectal Cancer Screening Stacy B. Menees, John Inadomi, Sheryl Korsnes, Grace Elta Background: The preference of women patients for women physicians has been shown in many specialties. In the primary care setting, it has not been examined whether women undergoing colonoscopy for colorectal cancer(CRC) screening prefer a female endoscopist nor have the reasons for this preference been studied. Aim: To examine barriers to CRC screening among women in a primary care setting. We hypothesized that in women with a female physician gender preference, the lack of available women endoscopists is a significant barrier to colonoscopy. Methods: A questionnaire based on theoretical constructs from validated surveys and our previous work was administered to a prospective cohort of female patients awaiting primary care appointments at 4 offices. Patients reported sociodemographic characteristics, experiences with CRC, barriers to CRC screening, physician gender preference (and the reasons for it) and the strength of that preference. Results: 202 women patients aged 40-70 years completed the questionnaire. Average age was 53 years (SD68.1). Fourty-three percent preferred a female endoscopist of which 87% would be willing to wait >30 days for a female endoscopist; 14% would be willing to pay more for one. The most common reason (in 75%) for this gender preference was embarrassment. Univariate analysis revealed that the gender of the primary care physician (PCP), younger patient age, and if they were employed were predictors for preference for a female endoscopist. Of these variables, only female gender of the PCP (OR=3.40; 95% CI =1.59, 5.63), and employment (OR=2.53; 95% CI =1.21, 4.51) were predictors for female endoscopist preference by multivariate analysis. Barriers to colonoscopy included: disliked prep and test(31%), test too uncomfortable(18%), and CRC screening not relevant(13%). Five percent would not undergo a colonoscopy unless guaranteed a female endoscopist. The sole independent factor associated with adherence to screening was PCP recommendation (OR=2.93; 95% CI= 1.63-5.39). Conclusions: Women patients frequently prefer a female endoscopist and this preference is strong enough to delay the procedure and incur additional out of pocket expense. It is an absolute barrier to endoscopy in 5% of women. Interventions to improve adherence to CRC screening by women in the primary care setting must address this issue.
GASTROINTESTINAL ENDOSCOPY
P109
****612 Prospective Trial of Endoscopic Clips versus Combination Therapy in Upper GI Bleeding (PROTECCT—UGI Bleeding) John Saltzman, Shelley Walsh, Veruska Di Sena, Christopher Huang, Benjamin Merrifield, Lisa Strate, Rie Ookubo, David Carr-Locke Purpose: Combination therapy (CT) with injection of epinephrine and multi-polar electrocoagulation is a common endoscopic treatment for non-variceal UGI bleeding. Endoscopic hemoclips (HC) have theoretical advantages over CT and are also widely used as a primary treatment. The aim of this study is to compare the efficacy of HC versus CT in the treatment of acute non-variceal UGI bleeding. Methods: Adult outpatients and inpatients with UGI bleeding at a large university hospital were eligible to participate. Inclusion criteria were active non-variceal UGI bleeding, a non-bleeding visible vessel (NBVV), or active bleeding or a NBVV after the mechanical removal of an adherent clot. Patients were randomized to either HC or CT at the time of the decision to perform endoscopic therapy. All outcomes were assessed including rebleeding and death 30 days after initial therapy. Results: Forty-seven patients were studied: 26 patients randomized to HC and 21 to CT. The mean age, sex, ethnicity, comorbidity scores (Charleston Index), and use of NSAID or anticoagulant drugs were not different between the two groups. The most common endoscopic finding in both groups was an ulcer (81% HC and 86% CT patients). There were 22 patients with active bleeding (14 HC, 8 CT, p=0.55), 13 with a NBVV (6 HC, 7 CT), and 12 with an adherent clot (6 HC, 6 CT). Primary hemostasis was similar between the two groups with successful initial control of bleeding in 26/26 HC patients (100%) and 20/21 (95.2%) CT patients, p=0.45. The rebleeding rates were also similar in both groups: 4/26 HC patients (15.4%) versus 5/21 CT patients (23.8%), p=0.49. In the patients who rebled, no association was demonstrated between stigmata of hemorrhage and treatment arm, p=0.19. Overall for HC vs. CT patients the median length of hospital stay (4 6 2 vs. 4 6 0.8 days), median units of blood transfused (4 6 1 vs. 5 6 1 units), surgery rates (3 vs. 1 patient), and mortality (0 vs. 2 patients, none directly from GI bleeding) were not different. The median duration of endoscopic therapy was 20 6 2 minutes in the HC group vs. 20 6 5 minutes in the CT group, p=0.45. One patient (CT group) developed aspiration pneumonia as a complication of the endoscopic procedure. Conclusions: In this first prospective, randomized controlled trial of endoscopic hemoclips versus combination therapy in nonvariceal upper GI bleeding, the efficacy and complications to date of the two treatment modalities are similar.
VOLUME 59, NO. 5, 2004
****613 Endoscopic Variceal Ligation (EVL): The More, the Merrier? A Prospective Multicenter Randomized Controlled Study Jose D. Sollano Jr., Frederick T. Dy, Stephen N. Wong, Melchor M. Chan, Albert E. Ismael, Jose A. Tan, Evan G. Ong, Marvin D. Basco Background: Bleeding from ruptured esophageal varices is a catastrophic sequela of portal hypertension and is associated with a high mortality rate. Endoscopic variceal ligation (EVL) can completely obliterate varices and is currently the treatment of choice for acute variceal bleeding. Due to its low complication rate and ease of deployment, ligating as many varices as possible in a single session may improve outcomes and cut costs of therapy. The objective of this study is to determine the outcome of deploying more than 6 bands vs. 6 bands or less per session of EVL. Methods: All patients with evidence of active or recent bleeding from esophageal varices with no history of previous EVL, sclerotherapy or intake of beta-blockers for the past 6 months were randomized into two groups (Group 1 >6 bands; Group 2 <6 bands; Six Shooter Saeed Multi-band Ligator, WilsonCook, Winston-Salem, NC.). Repeat endoscopies were done at day 10 and at 7-day intervals thereafter until all varices were obliterated. Patients did not receive additional medications. Follow-up endoscopies were scheduled every 6 months. Primary endpoints were complete variceal obliteration and rebleeding. Mortality was considered as a secondary endpoint. Intent-to-treat analysis was done for all endpoints. Results: A total of 59 patients were randomized (Group 1=28; Group 2=31). Baseline characteristics were similar (p>0.05) although there was more alcoholic cirrhosis in Group 2 (Group 1=21.4% vs. Group 2=51.6%, p=0.03). Mean follow-up was 13.5 +/ÿ 9.7 mos. for all patients. Sessions needed for complete variceal obliteration were similar (Group 1=1.6 vs. Group 2=1.9, p=0.209). There was a failure to achieve complete obliteration in 18% and 13% of patients in Groups 1 and 2, respectively. There was no difference in both rebleeding (42.9% vs. 25.8% overall rebleeding; mean interval to rebleeding 13.6 vs. 20 mos.; 2-year rebleeding rate 65.3% vs. 31.8%; Group 1 vs. 2, p=0.182) and survival between the 2 groups (60.7% vs. 71% overall survival; 17.4 vs. 19.7 mos. mean survival; 40% vs. 69% 2-year survival rate; Group 1 vs. 2, p=0.439). On multivariate analysis, female sex was associated with higher rebleeding rates while Child-Pugh class, rebleeding and failure to obliterate the varices were associated with higher mortality rates. Conclusions: In EVL, deploying more than 6 bands per session does not reduce the sessions needed for variceal obliteration and offers no added advantage in reducing rebleeding and mortality rates.
**673 Women's Preference for Women Physicians Is a Barrier to Colorectal Cancer Screening Stacy B. Menees, John Inadomi, Sheryl Korsnes, Grace Elta Background: The preference of women patients for women physicians has been shown in many specialties. In the primary care setting, it has not been examined whether women undergoing colonoscopy for colorectal cancer(CRC) screening prefer a female endoscopist nor have the reasons for this preference been studied. Aim: To examine barriers to CRC screening among women in a primary care setting. We hypothesized that in women with a female physician gender preference, the lack of available women endoscopists is a significant barrier to colonoscopy. Methods: A questionnaire based on theoretical constructs from validated surveys and our previous work was administered to a prospective cohort of female patients awaiting primary care appointments at 4 offices. Patients reported sociodemographic characteristics, experiences with CRC, barriers to CRC screening, physician gender preference (and the reasons for it) and the strength of that preference. Results: 202 women patients aged 40-70 years completed the questionnaire. Average age was 53 years (SD68.1). Fourty-three percent preferred a female endoscopist of which 87% would be willing to wait >30 days for a female endoscopist; 14% would be willing to pay more for one. The most common reason (in 75%) for this gender preference was embarrassment. Univariate analysis revealed that the gender of the primary care physician (PCP), younger patient age, and if they were employed were predictors for preference for a female endoscopist. Of these variables, only female gender of the PCP (OR=3.40; 95% CI =1.59, 5.63), and employment (OR=2.53; 95% CI =1.21, 4.51) were predictors for female endoscopist preference by multivariate analysis. Barriers to colonoscopy included: disliked prep and test(31%), test too uncomfortable(18%), and CRC screening not relevant(13%). Five percent would not undergo a colonoscopy unless guaranteed a female endoscopist. The sole independent factor associated with adherence to screening was PCP recommendation (OR=2.93; 95% CI= 1.63-5.39). Conclusions: Women patients frequently prefer a female endoscopist and this preference is strong enough to delay the procedure and incur additional out of pocket expense. It is an absolute barrier to endoscopy in 5% of women. Interventions to improve adherence to CRC screening by women in the primary care setting must address this issue.
GASTROINTESTINAL ENDOSCOPY
P109