Endoscopic Ultrasound in Patients Over 80 Years Old

Transcription

1 DOI /s ORIGINAL ARTICLE Endoscopic Ultrasound in Patients Over 80 Years Old Tan Attila Douglas O. Faigel Received: 11 January 2011 / Accepted: 8 April 2011 Ó Springer Science+Business Media, LLC 2011 Abstract Background The data on the safety and utility of EUS in patients over 80 years of age is limited. Objective We investigated the indications, feasibility, safety, and clinical utility of EUS in the management of GI, pancreatobiliary, and mediastinal diseases. Design Retrospective study. Setting A tertiary referral university hospital. Patients Consecutive patients C80 years of age referred over a 9-year period for EUS evaluation. Results A total of 265 EUS scans were performed in 232 patients with a mean age of 83.8 years. The indications for pancreatobiliary EUS were to evaluate a pancreatic mass (n = 60), pancreatic cyst (n = 18), pancreatitis (n = 10), dilated CBD in the setting of jaundice and/or biliary stricture (n = 20), dilated CBD with no jaundice and/or biliary stricture (n = 20). The indications for luminal gastrointestinal EUS were esophageal lesions (n = 21), gastric lesions (n = 30), duodenal lesions (n = 14), rectal lesions (n = 22), and gastrointestinal subepithelial lesions (n = 28). The indications for mediastinal EUS were mass/ lymphadenopathy (n = 14). EUS-guided FNA (EUS-FNA) was performed in 95 (35.8%) cases and results were consistent or suspicious for a malignancy in 62 cases (65.2%). Endoscopic mucosal resection (EMR) was performed in 17 cases (6.41%) on the same session following endosonographic evaluation. The procedure was successful in all patients with no complications related to sedation, EUS, or T. Attila D. O. Faigel (&) Division of Gastroenterology and Hepatology, Oregon Health and Science University, 3181 SW Sam Jackson Park Rd (L462), Portland, OR , USA faigeld@ohsu.edu EUS-FNA encountered. One patient had perforation following EMR. Limitation Retrospective study. Conclusions EUS and EUS-FNA are feasible and safe and have a significant impact on the management of GI, pancreatobiliary and mediastinal diseases in extreme elderly. Keywords Endoscopic ultrasound Fine needle aspiration Elderly Introduction As life expectancy increases, the number of older individuals has been increasing in many countries. As a result, this group represents a growing proportion of the total population. According to a US Department of Health and Human Services report in 2007, the older population (65? years) numbered 37.3 million in 2006 [1]. By 2030, the older population is expected to reach 71.5 million (more than twice the number in 2000). As a result of this trend, gastroenterologists are taking care of and performing endoscopic procedures on an increasing number of elderly patients. The role of endoscopic ultrasound (EUS) is well established in gastrointestinal, pancreatobiliary, and lung cancer diagnosis and staging [2]. The incidence of cancer in the population over 65 years is ten times higher than in those younger than 65 years and the cancer death rate is 16 times greater in patients over 65 years compared to younger patients [3]. Therefore, EUS is expected to play an important role in the care of elderly patients. Patients over 80 years old represent a challenging subgroup of the elderly. In these patients it is important to

2 balance life expectancy, quality of life, and safety concerns against the potential benefits of performing EUS. Unfortunately, there is a lack of data on EUS in the extreme elderly. The aim of this study is to describe clinical and demographic characteristics, indications, feasibility, safety, and impact of EUS in patients over 80 years of age from a tertiary referral center. Materials and Methods Patients All EUS procedures performed from March 1998 to March 2007 at a tertiary referral center (Oregon Health and Science University, Portland, Oregon) were reviewed. Patients 80 years of age or older were identified. The demographics, EUS indications, procedural information (instruments, sedation, American Society of Anesthesiology (ASA) classification, duration of the procedures, findings, interventions, outcome and complications) were retrospectively analyzed. Procedures were performed on both an inpatient and outpatient basis. Equipment and Procedure All EUS procedures were performed by an experienced endosonographer. All patients had a radial endosonographic evaluation (GF-UM160; Olympus, Melville, NY). Curvilinear echoendoscopes (GF-UC140P AL5; Olympus, Melville, NY; or FG36UX: Pentax Precision Medical Orangeburg, NY) were only used after radial endosonographic examination when clinically appropriate to perform EUS-guided FNA (EUS-FNA) or EUS-guided fine needle injection (EUS-FNI). A 19- or 22-gauge needle occluded with a stylet (Wilson Cook Inc, Winston-Salem, NC) was used for EUS-FNA and EUS-FNI. Color Doppler examination was performed before EUS-FNA or EUS-FNI to exclude interposed vascular structures. Intravenous antibiotics were routinely administered only in case of cystic lesion sampling or in patients with jaundice. Ciprofloxacin was the preferred antibiotic. Prophylactic antibiotics were not administered to patients undergoing EUS-FNI. High-frequency ultrasound miniprobe evaluation was performed with a 20-MHz Olympus miniprobe (Olympus Co., Ltd., Tokyo, Japan). A cytopathologist was present during each EUS-FNA procedure. Three different types of sedation were used: moderate (conscious) sedation, deep sedation, and general anesthesia. Fentanyl, midazolam, promethazine, and propofol were used for sedation. Sedation was administered by the endoscopist in cases of moderate sedation and by an anesthesiologist or certified registered nurse anesthetist in case of deep sedation or general anesthesia. The type of sedation was chosen based on comorbidities and previous experience with sedation. Procedures were performed with patients in the left lateral decubitus position. Results Patient Population Over the 9-year period, 265 (8.26%) of 3,205 procedures were performed on 232 patients who were 80 years old or older. Twenty-two patients had more than one procedure. There were 140 (60.3%) female and 92 (39.7%) male patients. Patients ages ranged from 80 to 97 years, with a mean age (standard deviation [SD]) of 83.8 (1.4) years. Anesthesia The ASA classification of each patient for every procedure was determined; two patients were classified as ASA 1, 180 patients were classified as ASA 2, and 83 patients were classified as ASA 3. The majority of the EUS procedures [224 (84.52%)] were performed under conscious sedation. Twenty-eight (10.56%) procedures were performed with the patients under general anesthesia with endotracheal intubation or deep sedation and 13 procedures (4.90%) were performed with no sedation. Average dosage of medications used for conscious sedation were: fentanyl (mean ± standard deviation) 101 ± 43.1 lg (range: ), midazolam 4.6 ± 1.7 mg (range: 1 10). Seventeen patients received promethazine 17.1 ± 8.7 mg (range: ). Procedure Information Two hundred and forty-three (91.7%) upper and 22 (8.3%) lower EUS examinations were performed. Ten patients (3.7%) underwent high-frequency ultrasound miniprobe evaluation. The average procedure (SD) duration was 52 (±18) min. One hundred and twenty-eight patients (48.3%) underwent EUS to evaluate pancreatobiliary pathologies. Indications for pancreatobiliary EUS were to evaluate a pancreatic mass (n = 60), pancreatic cystic lesion (n = 18), pancreatitis (n = 10), dilated common bile duct in the setting of jaundice and/or biliary stricture (n = 20), dilated CBD with no evidence of biochemical cholestasis and/or biliary stricture (n = 20). Eighty-seven patients (30.8%) underwent EUS for luminal gastrointestinal tract pathologies; esophageal (n = 21), gastric (n = 30), duodenal (n = 14), and rectal (n = 22). Twenty-eight patients (10.5%) underwent EUS for gastrointestinal subepithelial pathologies. Fourteen

3 patients (5.2%) underwent EUS for evaluation of mediastinal pathologies. Eight patients (3%) had EUS evaluation for further evaluation of other pathologies. Ninety-five patients (35.8%) had EUS-FNA. No EUS-associated complications occurred. One patient had a perforation complicating EMR of subepithelial lesion (GIST) (see below). Sixty patients were referred for evaluation of pancreatic mass lesions (Table 1). Endosonographic evaluation revealed solid pancreatic mass lesions (n = 29), ampullary mass lesions (n = 13), pancreatic cystic lesions (n = 8), pancreatitis (n = 2), fatty infiltration (n = 2), and normal pancreatic parenchyma (n = 6). Among 29 patients with pancreatic mass lesions; EUS-FNA was performed on the pancreatic mass (16 patients), on the pancreatic mass and associated lymph nodes (eight patients) and on a liver mass (one patient). Cytopathological evaluation of FNA sample of pancreatic mass ± lymph nodes revealed adenocarcinoma (n = 17), suspicion for malignancy (n = 2), renal cell carcinoma (n = 1), and no findings for malignancy (n = 4). Cytopathological evaluation of FNA sample obtained of the liver mass revealed adenocarcinoma and lymph node revealed neuroendocrine tumor. Among 13 patients with an ampullary mass, one patient had EUS-FNA of the ampullary mass, which revealed adenocarcinoma and two patients had EUS-FNA of lymph nodes which were reactive. EUS-FNA was performed on all eight patients with a pancreatic cystic lesion. One patient was diagnosed with a mucinous adenocarcinoma (23 mm in diameter). A carcinoembryonic antigen (CEA) level could not be measured due to an inadequate amount of fluid obtained to run the test. Another patient who was diagnosed with a mucinous neoplasm with dysplastic features (90 mm in diameter) had a CEA level of 2,887 ng/dl. In the remaining six patients with pancreatic cysts, EUS-FNA samples were negative for a malignancy. Eighteen patients were referred with a primary indication to evaluate pancreatic cystic lesions (Table 2). In 13 patients, cysts did not have any endosonographic evidence of main duct connection. Eleven EUS-FNA performed revealed mucin (n = 3), adenocarcinoma (n = 1), atypical cells (n = 1), pseudocyst (n = 1), and no malignancy (n = 5). CEA was below 192 ng/ml in five patients and above 192 ng/ml in two patients. CEA could not be measured in the rest of the patients due to the inadequate fluid sample. Two patients were found to have cystic lesions with main pancreatic duct connection. EUS-FNA revealed Table 1 Endoscopic ultrasound performed for the evaluation of pancreatic mass lesions (n = 60) FNA fine needle aspiration; LN lymph node; AdenoCA adenocarcinoma Findings No. of patients FNA FNA findings Pancreas mass (mass) AdenoCA (17) 8 (mass?ln) Suspicious for malignancy (2) Renal cell carcinoma (1) Negative for malignancy (4) 1 (liver mass) AdenoCA (1) 1 (LN) Neuroendocrine tumor (1) Ampullary mass 13 1 (mass) AdenoCA (1) 2 (LN) Reactive (2) Cystic lesion 8 8 (mass) Mucinous AdenoCA (1) Mucinous neoplasm with dysplastic features (1) Negative for malignancy (6) Pancreatitis 2 Fatty infiltration 2 No mass 6 Table 2 Endoscopic ultrasound evaluation of pancreatic cystic lesions MPD main pancreatic duct; CMPD communicating with main pancreatic duct; NCMPD non-communicating with main pancreatic duct Findings No. of patients FNA FNA findings CEA Isolated MPD dilation 3 (17, 18, 4 mm) Cysts CMPD 2 (34, 42 mm) 2 Mucin: 2 139, 265 Cysts NCMPD 13 (average 24.6 mm) 11 AdenoCA: 1 [192 : 2 Atypical cells 1 Mucin: 3 \192 : 5 Pseudocyst: 1 Negative for malignancy 5

4 mucin and CEA levels of 139 and 265 ng/ml. Three patients were found to have an isolated dilation of the main pancreatic duct. Ten patients were referred for the evaluation of pancreatitis. EUS criteria for chronic pancreatitis described by Catalano et al. [5] were used. One patient fulfilled endosonographic criteria for chronic pancreatitis. Among nine patients with endosonographically normal pancreatic parenchyma, incidental findings were pancreatic cysts in two patients, pancreatic duct stone without other inflammatory changes in one patient, and dilated CBD secondary to ampullary stenosis. Twenty patients were referred for the evaluation of dilated common bile duct (CBD) in the setting of jaundice and/or stricture (Table 3). Eight patients were found to have a pancreatic mass. Seven patients who had EUS-FNA were found to have adenocarcinoma (n = 5), suspicion of malignancy (n = 1), and no malignancy (n = 1). One patient with a pancreatic mass did not have EUS-FNA due to agitation. CBD mass lesions were noted in five patients. EUS-FNA performed on these patients revealed adenocarcinoma (n = 1), metastatic adenocarcinoma (n = 1), atypical cells (n = 1), and no malignancy (n = 2). In the remaining patients, the cause of dilated CBD was found to be pancreatitis (n = 3), ampullary mass (n = 2), retroperitoneal mass (n = 1), and CBD stone (n = 1). Twenty patients were referred for the evaluation of dilated CBD without any biochemical evidence of cholestasis or biliary stricture. Average endosonographic diameter of CBD was 13.7 mm (±3.4 mm). Eight patients had a prior history of cholecystectomy. Seven patients had endosonographic evidence of CBD stones and sludge and were referred to ERCP. The remaining 13 patients had no endosonographic evidence of biliary pathology and the presumed cause of CBD dilation was thought to be secondary to benign ampullary stenosis. Fourteen patients were referred for endosonographic evaluation of mediastinal mass lesions with an average size of 32.7 mm (±15.5 mm) (Table 4). Thirteen EUS-FNA were performed diagnosing adenocarcinoma (n = 3), nonsmall cell lung cancer (n = 3), small cell lung cancer (n = 2), renal cell carcinoma (n = 1), and no evidence of malignancy (n = 4). Three patients with negative EUS- FNA were subsequently diagnosed with locally advanced thymoma, small cell lung cancer on mediastinoscopy, and thymoma on bronchoscopy. Table 3 EUS evaluation of dilated CBD in the setting of jaundice and/or stricture CBD common bile duct; FNA fine needle aspiration; AdenoCA adenocarcinoma; LN lymph node Findings No. of patients Size FNA FNA findings Pancreatic mass 8 29(±7 mm) 5 (mass) AdenoCA (5) 1 (mass? LN) Suspicious for malignancy (1) 1 ascites Negative for malignancy (1) CBD mass 5 20 (±7 mm) 3 (mass) Negative for malignancy (2) 1 (LN) Metastatic adenoca (1) 1 (mass? LN) AdenoCA (1) Atypical cells (1) Pancreatitis 3 1(mass? LN) Pancreatitis (1) 1 (LN) Negative for malignancy (1) Reactive changes (1) Ampullary mass 2 Retroperitoneal mass 1 CBD stone 1 19 Table 4 Endoscopic ultrasound for the evaluation of mediastinal lesions AdenoCA adenocarcinoma; SCLCa small cell lung cancer; NSCLCa non-small cell lung cancer; FNA fine needle aspiration No. of patients FNA FNA findings Intervention AdenoCA (3) SCLCa (2) NSCLCa (3) Renal cell carcinoma (1) Negative for malignancy (4) Mediastinoscopy (2) Locally advanced thymoma SCLCa Bronchoscopy (1) Thymoma

5 Table 5 Esophageal cases Indications No. of patients FNA Outcome Endoscopic Rx Unable to traverse AdenoCA 13 1 LN Neg: 1 EMR 4 2 SCC 5 4 LN Neg: 2 Mets: 2 Dysphagia strictures 2 1 LN Neg: 1 Barrett s esophagus HGD 1 RFA 1 AdenoCA adenocarcinoma; SCC squamous cell carcinoma; HGD high-grade dysplasia; Neg negative; EMR endoscopic mucosal resection; RFA radiofrequency ablation; LN lymph node; N number of Table 6 Endoscopic ultrasound evaluation of gastric lesions Indications No. of patients FNA FNA findings EMR EMR findings Gastric cancer 14 2 AdenoCA (2) Post-EMR f/u 5 3 Inflammatory polyp (1) Lipoma (1) IMD (1) GJA thickening 1 1 (LN) AdenoCA Fundus mass 2 FU 1 Gastric wall thickening 6 (2 normal) Non-healing AU 1 IMD intestinal metaplasia with dysplasia; AdenoCA adenocarcinoma; FNA fine needle aspiration; EMR endoscopic mucosal resection; f/u follow-up; GJA gastrojejunal anastomosis; LN lymph node; FU fundus ulcer; AU antrum ulcer Twenty-one patients were referred for endosonographic evaluation of esophageal lesions (Table 5); adenocarcinoma (n = 13), squamous cell carcinoma (n = 5), stricture (n = 2), and Barrett s esophagus with high-grade dysplasia (n = 1). In two patients, the echoendoscope could not traverse the esophageal narrowing caused by adenocarcinoma. EUS-FNA of five patients with mediastinal lymphadenopathy revealed squamous cell carcinoma metastasis in two patients and no malignancy in the remaining three patients. Following EUS evaluation, four patients with esophageal adenocarcinoma and one patient with Barrett s esophagus with high-grade dysplasia underwent EMR (n = 4) and radiofrequency ablation (n = 1). Thirty patients were referred for endosonographic evaluation of gastric lesions (Table 6). Fourteen patients had EUS for gastric cancer. Two of them subsequently underwent EMR that revealed adenocarcinoma. Five patients had EUS as post-emr follow-up. Three of them subsequently had EMR that revealed an inflammatory polyp, a lipoma and intestinal metaplasia with dysplasia. A patient with a prior history of gastric adenocarcinoma had an EUS for evaluation of gastrojejunal thickening on computerized tomography. A subsequent EUS-FNA of a lymph node showed adenocarcinoma recurrence. The remaining gastric EUS examinations were done for the evaluation of gastric wall thickening (n = 6), fundic mass (n = 2), fundic ulcers (n = 1), and non-healing antral ulcer (n = 1). No endosonographic evidence of a fundic mass was noted. No evidence of gastric wall thickening was noted in four patients. The other patient had an ulcer with wall thickening likely secondary to inflammatory changes. Other patients had normal endosonographic findings. Fourteen patients with duodenal lesions were referred for endosonographic evaluation. Ten patients with duodenal adenoma had high-frequency ultrasound miniprobe evaluation. In nine patients, duodenal adenomas were subsequently ablated with argon plasma coagulation. The other duodenal adenoma was removed with EMR. A patient with a suspicion of duodenal varix was found to have a periduodenal cyst. A patient with a suspicious periduodenal extrinsic compression at previous endoscopic evaluation had normal endosonographic findings. The remaining two patients were referred for endosonographic evaluation of duodenal adenocarcinoma and duodenal polyp (adenoma). Twenty-two patients were referred for endosonographic evaluation of rectal lesions; rectal mass (n = 17), rectal malignancy recurrence (n = 4), and perirectal mass (n = 1). An incidental subepithelial lesion was noted at the endosonographic evaluation for a rectal cancer recurrence follow-up. EUS-FNA of this lesion revealed GIST.

6 Table 7 Endoscopic ultrasound evaluation of subepithelial lesions Locations No. of patients FNA FNA findings EMR EMR findings Bx Stomach 17 7 GIST (3) 4 Leiomyoma (2) SMN (1) Foveolar polyp (1) Lipoma (1) GIST (1) SCN (2) GEJ 6 1 SMN (1) 3 BE HGD (1) NL (1) Lymphangioma (1) Duodenum 3 1 Carcinoid (1) GIST (1) Esophagus 2 1 B cell lymphoma (1) GIST gastrointestinal stromal tumor; SMN smooth muscle neoplasm; SCN spindle cell neoplasm; NL normal; BE HGD Barrett s esophagus with high-grade dysplasia; GEJ gastroesophageal junction EUS-FNA of a perirectal mass revealed abundant anucleate squamous cells with no evidence of malignancy. These findings raised the possibility of epidermal inclusion cyst. Twenty-eight patients underwent endosonographic evaluation for subepithelial lesions (Table 7) [gastric (n = 17), gastroesophageal junction (n = 6), duodenal (n = 3), and esophageal (n = 2)]. EUS-FNA performed on seven patients with gastric lesions revealed GIST (n = 3), spindle cell neoplasm (n = 2), smooth muscle neoplasm (n = 1), and lipoma (n = 1). Four patients with gastric lesions subsequently underwent EMR, which revealed two leiomyomas, one foveolar polyp, and one GIST. The EMR that was done to remove gastric GIST was complicated with perforation. EUS-FNA that was done on a gastroesophageal junction lesion revealed smooth muscle neoplasm. Three EMRs of gastroesophageal lesions revealed a Barrett s esophagus with high-grade dysplasia, a lymphangioma, and a normal histopathology. An EMR of duodenal lesion revealed a carcinoid tumor. A duodenal GIST was diagnosed with conventional biopsy. An esophageal B cell lymphoma was diagnosed with EUS- FNA. Endosonography of the other esophageal subepithelial lesion revealed a duplication cyst. Four patients who were referred for endosonographic evaluation of suspected upper GI tract pathology had unremarkable EUS findings (a patient with signet ring cell adenocarcinoma on breast biopsy, a patient with abdominal pain, a patient with weight loss, and a patient with lymphoma). Two patients with abnormal liver function tests were endosonographically evaluated for lymphadenopathy. EUS-FNA of a periportal lymph node of a patient with lung mass revealed lung cancer metastasis. A patient with a history of colon cancer was referred for porta hepatis lymphadenopathy. EUS evaluation did not reveal any pathologic lymph nodes but endosonographic findings were consistent with chronic pancreatitis. Discussion Although the role of EUS is well established in adult GI and pancreatobiliary diseases as well as lung cancer diagnosis and staging, there is no specific knowledge in the literature regarding the role of EUS in patients over 80 years of age. The current study illustrates the experience of endoscopic ultrasound at a tertiary referral center evaluating various pancreatobiliary, GI tract and mediastinal lesions in extreme elderly. Given the aging population, gastroenterologists will perform more advanced diagnostic procedures in extremes of ages. Prior studies have reported on the indications and complications of EGD, colonoscopy, and ERCP in patients 80 or older [6]. These studies have generally found these procedures to be safe and effective, and the complications not higher than those in younger patients. Ours is the first study of EUS in this cohort. According to the US Census Bureau s 2007 American Community Survey, individuals over 80 make up 3.6% of the population [4]. However, in our cohort, 8.26% of the EUS procedures were performed on the same age group. This discrepancy is likely due to the higher incidence of malignancies in this age group. Not surprisingly, the vast majority of EUSs were done due for a suspicion of malignancy. EUS-FNA results were consistent with a malignancy or suspicious for a malignancy in 62 out of 95 (65.2%) EUS-FNA performed. Endoscopic mucosal resections were performed in 17 patients (6.41%) upon endosonographic evaluations. All of the EMR results allowed accurate diagnosis and staging of

7 the underlying pathologies. In the extreme elderly seen at our institution, the primary role of EUS was cancer diagnosis and staging. In these older patients with comorbidities, treatment may be more limited, but a diagnosis is important to allow for appropriate counseling. Although only half of the EUS procedures (48.3%) were performed for the evaluation of the pancreatobiliary system in patients over 80, more than two-thirds (67.3%) of the EUS-FNAs were performed to evaluate pancreatobiliary pathologies. Of the 64 patients who underwent pancreatobiliary EUS-FNA, 40 (62.5%) had cytopathologic results consistent with malignancy or suspicious for malignancy. Among the 19 patients with no cytopathologic evidence of malignancy, ten EUS-FNA were obtained from pancreatic cystic lesions. As expected, the yield to demonstrate malignant cells in the fluid obtained from pancreatic cysts is low. Therefore, the presence of mucin and cyst CEA levels plays an important role in risk stratification of pancreatic cystic lesions. The major limitation of this study is the retrospective design. Since a good portion of our practice is based on referral from other institutions, information on surgical pathology as well as clinical outcome of our patients is scarce. Therefore, we cannot comment on the correlation of EUS-FNA results with surgical pathologies. The prereferral evaluations were not standardized. We cannot exclude a referral bias as to why these elderly patients were referred but others may not have been. All patients tolerated the procedure well. No EUS or EUS-FNA-related complications occurred. One patient had a perforation during the EMR that was done following endosonographic evaluation, which was not considered an EUS-related complication. In conclusion, EUS and EUS-FNA are safe and have a significant impact on the management of pancreatobiliary, gastrointestinal, and mediastinal diseases in patients over 80 years of age. EUS allows one to avoid the more-invasive and higher-risk procedures. The primary role appears to be for the diagnosis of malignancy. Conflict of interest The authors (Tan Attila, MD; Douglas O. Faigel, MD) declare that they have no conflicts of interest. References 1. A profile of Older Americans. Administration on aging. US Department of Health and Human Services. prof/statistics/profile/2007/2007profile.pdf; LeBlanc JK, DeWitt J, Sherman S. Endoscopic ultrasound: how does it aid the surgeon? Adv Surg. 2007;41: Berger NA, Savvides P, Koroukian SM, et al. Cancer in the elderly. Trans Am Clin Climatol Assoc. 2006;117: Catalano MF, Lahoti S, Geenen JE, Hogan WJ. Prospective evaluation of endoscopic ultrasonography, endoscopic retrograde pancreatography, and secretin test in the diagnosis of chronic pancreatitis. Gastrointest Endosc. 1998;48: Qureshi WA, Zuckerman MJ, Adler DG, et al. ASGE guideline: modifications in endoscopic practice for the elderly. Gastrointest Endosc. 2006;63: