Maurits J. Wiersema

Endosonography-guided fine-needle aspiration biopsy: diagnostic accuracy and complication assessment.

BACKGROUND & AIMS Endosonography-guided fine-needle aspiration biopsy (EUS-FNA) permits cytological confirmation of EUS findings. A multicenter prospective evaluation of EUS-FNA for primary diagnosis, staging, and/or follow-up purposes was undertaken. METHODS EUS-FNA was performed in 457 patients with 554 lesions. Clinical (n = 218) or histopathologic (n = 256) confirmation was available in 192 lymph nodes, 145 extraluminal masses, 115 gastrointestinal wall lesions, and 22 cystic lesions. RESULTS EUS-FNA sensitivity, specificity, and accuracy was 92%, 93%, and 92% for lymph nodes, 88%, 95%, and 90% for extraluminal masses, and 61%, 79%, and 67% for gastrointestinal wall lesions, respectively. The sensitivity and accuracy for lymph nodes and extraluminal masses was superior to that for gastrointestinal wall lesions. When EUS-FNA was compared with EUS size criteria in lymph node evaluation, specificity (93% vs. 24%) and accuracy (92% vs. 69%) were superior, whereas sensitivity (92% vs. 86%) was similar. The accuracy of EUS-FNA in patients with previously failed biopsy procedures was 81% (73 of 90). Five nonfatal complications occurred for a rate of 0.5% (95% confidence interval, 0.1%-0.8%) in solid lesions vs. 14% (95% confidence interval, 6%-21%) in cystic lesions. CONCLUSIONS EUS-FNA accurately and safely evaluates solid peri-intestinal lesions and improves lymph node staging accuracy.

Endosonography-guided fine needle aspiration biopsy in the evaluation of pancreatic masses

OBJECTIVES:Diagnosis of pancreatic tumors can be problematic. This study aimed to determine the performance of endoscopic ultrasound-guided fine needle aspiration biopsy (EUS FNA) in pancreatic malignancy when prior biopsies performed by CT guidance or ERCP were negative.METHODS:A total of 185 patients with known or suspected pancreatic masses were prospectively evaluated with EUS FNA. Before EUS FNA, all patients were evaluated with abdominal CT (61 with CT-guided biopsy) and 91 with ERCP (41 had brushings or biopsy).RESULTS:EUS had greater sensitivity than CT in detecting a mass (99% vs 57%, p < 0.0001). In 58 patients with negative CT-guided biopsies, EUS FNA had 90% sensitivity for malignancy, 50% specificity for benign disease and 84% accuracy. Similarly, in 36 patients with negative ERCP tissue sampling, results for EUS FNA were 94%, 67% and 92%, respectively. Complications were mild and infrequent (0.5%).CONCLUSION:EUS FNA of pancreatic masses safely and accurately diagnoses pancreatic malignancy when prior biopsy techniques have been unsuccessful.

Endosonography-guided celiac plexus neurolysis

BACKGROUND We have evaluated the safety and efficacy of performing endosonography-guided celiac plexus neurolysis (EUS CPN) in patients with pain due to intra-abdominal malignancies. METHODS Thirty patients with upper abdominal pain requiring narcotic analgesia and suspected or known intra-abdominal malignancy were selected for EUS CPN. This group included 25 patients with pancreas carcinoma and 5 patients with intra-abdominal metastases. Using the linear array ultrasound endoscope and a prototype needle catheter, transgastric injection of the celiac plexus with bupivacaine and 98% dehydrated absolute alcohol was accomplished. RESULTS Pain scores were significantly lower compared with baseline at 2, 4, 8, and 12 weeks after EUS CPN (median follow-up: 10 weeks). At these follow-up intervals, 82% to 91% of patients required the same or less pain medication and 79% to 88% of patients had persistent improvement in their pain score. Comparison of patients with TXNXM1 versus TXNXMO pancreatic carcinoma revealed higher initial pain scores (7.9 +/- 1.92 versus 5.8 +/- 2.0, p = .02) and a greater decline in pain scores (decrease of 6.1 +/- 3.1 versus 4.8 +/- 2.0, p = .004). Complications were minor and consisted of transient diarrhea in four patients. CONCLUSION EUS CPN is a safe and effective means for improving pain control in patients with intra-abdominal malignancy. The technique may be performed as an outpatient at the same setting as the EUS staging examination.

A prospective, controlled assessment of factors influencing acceptance of screening colonoscopy

OBJECTIVES:Medicare beneficiaries now have access to screening colonoscopy (SC). For colon cancer screening to be fruitful, SC must become more acceptable to a broad segment of this population. However, we currently lack knowledge of which aspects of SC have an impact on patient acceptance. The aims of this study were: 1) to identify the features of SC that are most important in deterring participation, and 2) to prioritize and to compare the perceptions of never-screened individuals with those of individuals previously screened for colon cancer.METHODS:Questionnaires were distributed to 300 outpatients at Mayo Clinic, Rochester (150 never-screened patients; 150 previously screened patients). The survey instrument addressed domains of the Health Belief Model and colon cancer risk perception. Patients ranked the three most important barriers to SC and answered general knowledge questions on colon cancer.RESULTS:Response rates of never-screened (84%) and screened (88%) patients were similar. Never-screened patients were less likely to have a regular primary physician (80% vs 95%, p = 0.0003) and were less likely to have undergone a prior screening mammography (87% vs 96% of women, p = 0.02) compared with screened patients. The four most reported deterrents to SC (“volume of bowel preparation,” “ adequate analgesia,” “ no recommendation from primary physician,” and “embarrassment”) were ranked similarly by both groups. Never-screened patients had less understanding of the incidence and treatment outcomes of colon cancer.CONCLUSION:Colon cancer screening behavior seems to be associated with having a regular primary physician, as well as other cancer screening behaviors. Knowledge of colon cancer is the most reliable discriminator of prior screening status. There does not seem to be any difference in the preferences expressed by never-screened and screened patients with respect to the aspects of colonoscopy that they find objectionable.

Endosonography-guided cholangiopancreatography

BACKGROUND Detailed imaging of the common bile duct and main pancreatic duct is possible with endosonography. Utilizing a custom manufactured flexible needle, we have developed a technique of performing endosonography-guided cholangiopancreatography (EGCP). METHODS Of 205 patients undergoing ERCP, complete ductography was not possible in 11 patients. Employing a linear scanning echoendoscope in conjunction with a 4 cm, 22 to 23 gauge aspiration needle, transduodenal cholangiography (n = 10) or transgastric pancreatography (n = 1) was attempted. RESULTS Successful ductography was possible in 8 of 11 patients (EGCP success 73% vs ERCP 0%, p < .001, Fishers exact test). In 5 patients, abnormalities identified on EGCP subsequently led to repeat ERCP with precut sphincterotomy. In all of these cases 100% agreement was found between EGCP and ERCP findings. One postprocedure case of pancreatitis occurred in a patient who underwent EGCP at the same setting as the failed ERCP. No early or late complications occurred in the patient group with EGCP performed at a separate setting. CONCLUSION EGCP allows an alternative method for obtaining cholangiopancreatography in those patients in whom ERCP is unsuccessful. Further studies are necessary to define the safety and success rate of this new procedure.

Utility of EUS in the evaluation of cystic pancreatic lesions.

BACKGROUND Preoperative differentiation of benign and malignant/potentially malignant pancreatic cystic lesions is problematic. Data to support the role of EUS and EUS-guided fine-needle aspiration (EUS-FNA) are limited. This study assessed the sensitivity, specificity, and accuracy of EUS, cytopathology, and analysis of cyst fluid for pancreatic cystic lesions. METHODS Retrospectively, 111 consecutive patients were identified (54 men, 57 women; mean age 59 years, range 18-79 years) who underwent EUS from July 1997 to September 2000 because of known or suspected pancreatic cystic lesions based on CT or transabdominal US. Thirty-four patients (16 men, 18 women; mean age 55 years, 25-79 years) who underwent surgery formed the basis for this analysis. EUS diagnosis was compared with surgical pathology. Selected patients underwent EUS-FNA to obtain specimens for cytopathologic analysis and for determination of carcinoembryonic antigen levels. Based on surgical pathology, cysts were classified as benign (simple cyst, pseudocyst, serous cystadenoma) or malignant/potentially malignant (mucinous cystadenoma, intraductal papillary mucinous tumor, cystic islet cell tumor, cystic adenocarcinoma). RESULTS EUS-FNA with cytopathologic assessment of cyst fluid was performed for 18 of the 34 patients; carcinoembryonic antigen level was determined in 11 cases. For EUS, cytopathology, and carcinoembryonic antigen, sensitivity was, respectively, 91%, (p = 0.01 vs. cytology), 27%, and 28%; specificity was, respectively, 60%, 100%, and 25%; and, accuracy was, respectively, 82%, 55%, and 27%. The sensitivity of EUS in all 13 patients with cystic islet cell tumor, intraductal papillary mucinous tumor, or cystic adenocarcinoma was 100%. Combining EUS, cytopathology, and carcinoembryonic antigen results did not improve accuracy. There were no complications related to the EUS or EUS-FNA. CONCLUSIONS EUS alone is sensitive and accurate in identifying malignant/potentially malignant pancreatic cystic lesions. EUS-FNA to obtain specimens for cytopathologic analysis and determination of carcinoembryonic antigen levels, although safe, does not enhance diagnostic yield.

Safety and outcome of endoscopic snare excision of the major duodenal papilla

BACKGROUND The optimal management of adenoma of the major duodenal papilla is not established. Options include surgical excision, endoscopic ablative techniques, snare excision, and observation with periodic biopsies. The aims of this retrospective study were to determine the safety and outcome of snare excision of the papilla. METHODS Twenty-eight snare excisions of the papilla were performed in 26 patients. Sixteen had familial adenomatous polyposis. In 22 procedures, a minisnare was used, and in 6 cases a prototype snare was designed for excision of the papilla. Pancreatic stents were placed as a prophylactic measure at the discretion of the endoscopist (n = 10). RESULTS Histopathologically, resected tissue included 25 adenomas, 1 inflammatory polyp, 1 invasive malignancy, and 1 normal papilla. Immediate complications were minor bleeding (n = 2), mild pancreatitis (n = 4) and a duodenal perforation (n = 1). The presence (n = 10) or absence (n = 18) of a pancreatic stent did not correlate with subsequent pancreatitis (2 in each group, p = NS). Follow-up was available for 21 patients (median, 9 months; range, 2-32 months). Pancreatic duct stenosis at the papillectomy site resulted in pancreatitis in 2 patients (17%) at, respectively, 4 months and 24 months. Follow-up endoscopy revealed recurrent/residual adenomatous tissue in 2 (10%). CONCLUSIONS Snare excision of the major duodenal papilla was well tolerated. Most complications were mild except for a small duodenal perforation. Stenosis of the pancreatic duct orifice with pancreatitis may be a late complication.

Initial evaluation of the efficacy and safety of endoscopic ultrasound-guided direct ganglia neurolysis and block

BACKGROUND:Celiac plexus neurolysis and block are considered safe but provide limited pain relief. Standard techniques target the region of the celiac plexus but do not attempt injections directly into celiac ganglia. The recent recognition that celiac ganglia can be visualized by endoscopic ultrasound (EUS) now allows direct injection into celiac ganglia for neurolysis (CGN) and block (CGB).AIMS:To determine the safety and initial efficacy (at 2–4 wk) of direct ganglia injection in patients with moderate to severe pain secondary to unresectable pancreatic carcinoma or chronic pancreatitis.METHODS:An EUS database was reviewed to identify patients undergoing CGN and CGB. Data were retrieved from the medical records and phone follow-up.RESULTS:Thirty-three patients underwent 36 direct celiac ganglia injections for unresectable pancreatic cancer (CGN N = 17, CGB N = 1) or chronic pancreatitis (CGN N = 5, CGB N = 13) with bupivacaine (0.25%) and alcohol (99%) for CGN, or Depo-Medrol (80 mg/2 cc) for CGB. Cancer patients reported pain relief in 16/17 (94%) when alcohol was injected and 0/1 (00%) when steroid was injected. For chronic pancreatitis, 4/5 (80%) who received alcohol reported pain relief versus 5/13 (38%) receiving steroids. Thirteen (34%) patients experienced initial pain exacerbation, which correlated with improved therapeutic response (P < 0.05). Transient hypotension and diarrhea developed in 12 and 6 patients, respectively.CONCLUSIONS:Initial experience suggests that EUS-guided direct celiac ganglion block or neurolysis is safe. Alcohol injection into ganglia appears to be effective in both cancer and chronic pancreatitis. Prospective trials are needed to confirm the efficacy of this new approach.

EUS-guided trucut biopsy in establishing autoimmune pancreatitis as the cause of obstructive jaundice

BACKGROUND The diagnosis of autoimmune pancreatitis can be difficult and often requires a larger specimen than can be provided by FNA alone to determine if the tissue sample obtained with EUS trucut biopsy (TCB) is sufficient to allow adequate histologic review to establish the diagnosis of autoimmune pancreatitis. METHODS EUS TCB was performed in patients presenting with obstructive jaundice who were suspected of having autoimmune pancreatitis based on their clinical, laboratory and imaging studies. The charts were retrospectively reviewed to determine the feasibility of TCB. RESULTS Between August 2002 and June 2004, 3 patients with obstructive jaundice and suspected autoimmune pancreatitis (AIP) underwent EUS TCB. In each case, a diagnosis of pancreatic cancer also was considered, and surgical resection was the planned therapy before the patient underwent EUS TCB. Histologic review of the TCB specimens established the diagnosis of AIP in two patients and identified nonspecific changes of chronic pancreatitis in the third patient. EUS-guided FNA was performed in two of the 3 patients and failed to establish the diagnosis in either patient. Other than mild transient abdominal pain (n = 1), no complications were identified. CONCLUSIONS This preliminary study suggests that EUS TCB can safely establish the diagnosis of AIP. Doing so helps guide management and may help to avoid unnecessary surgery. Prospective studies are needed to verify these findings and to more clearly define the role of EUS TCB in these patients.

Effect of prophylactic main pancreatic duct stenting on the incidence of biliary endoscopic sphincterotomy-induced pancreatitis in high-risk patients

Pancreatitis is a common complication of endoscopic sphincterotomy. Cautery-induced papillary edema has been implicated as a possible cause. The objective of this study was to determine whether prophylactic stenting of the main pancreatic duct after endoscopic sphincterotomy in high-risk patients would reduce the incidence of pancreatitis. High-risk patients were defined as those with sphincter of Oddi dysfunction, small common bile duct diameter (< 10 mm), or those requiring pre-cut sphincterotomy. Patients were studied in a prospective fashion from October 1990 to April 1992 and were randomized to receive either a main pancreatic duct stent or no stent after biliary sphincterotomy. The stents were generally removed 10 to 14 days after placement. Fifty patients were randomized to the no-stent group and 48 patients were randomized to the stent group, but in five patients stent placement was unsuccessful. Pancreatitis occurred in 18% of patients in the no-stent group compared with 14% of patients in the stent group. Most cases of pancreatitis were mild, occurring in 10% of patients in the no-stent group and 12% of patients in the stent group. Moderate to severe pancreatitis occurred with an increased frequency in patients in the no-stent group (8%) compared with that in patients in the stent group (2%). Mean number of hospital days required to treat pancreatitis was 9.5 days in the no-stent group compared with 2.8 days in the stent group; however, none of these differences reached statistical significance. Small common bile duct diameter (< 6 mm) was found to be an independent risk factor for pancreatitis after endoscopic sphincterotomy.(ABSTRACT TRUNCATED AT 250 WORDS)