Search

BACKGROUND AND OBJECTIVE: The additional diagnostic value of endoscopic ultrasound-fine needle aspiration (EUS-FNA) over lymph node (LN) echofeatures alone in evaluating lymphadenopathy is unknown. The objectives of this study are (1) to prospectively evaluate the utility of EUS-FNA in evaluating mediastinal or peri-intestinal lymphadenopathy and to compare its yield to that of echofeatures alone and (2) to determine clinical and endosonographic features predictive of malignant involvement of LNs.METHODS: All consecutive patients who underwent EUS-FNA of a LN over a 22-month period were prospectively evaluated. Reference standard for final diagnosis included: surgery (n= 76), long-term clinical and/or imaging follow-up (n= 74), or death from disease (n= 26).RESULTS: One hundred and eighty-three EUS-FNAs of LNs were performed in 137 patients with no major complications. Locations of the biopsied LNs included 31% subcarinal, 21% celiac, 21% peripancreatic, 13% periesophageal, 4.4% aortopulmonary window, 3.2% perigastric, and 3.3% perirectal. Mean LN size was 20.5 mm (SD± 11.1)× 13.2 mm (SD± 7.97). The mean number of EUS-FNA passes was three (range 1–7). The sensitivity, specificity, PPV, and NPV of EUS-FNA of LNs were 98.3%, 100%, 100%, and 98.4%, respectively. EUS-FNA was more accurate compared to LN echofeatures alone (99.4%vs75.4%,p<0.001). Mediastinal LNs were 2.77 times less likely to be malignant as compared to other LN locations. In multivariable analysis, the number of LN echofeatures, site of LN, and patient's age were associated with malignant involvement (p= 0.001).CONCLUSIONS: EUS-FNA is superior to LN echofeatures in evaluating lymphadenopathy. Endosonographic LN features alone are particularly unreliable in the mediastinum, necessitating tissue confirmation. EUS-FNA can safely, reliably, and accurately sample mediastinal and peri-intestinal LNs obviating the need for more invasive testing or surgical intervention. [ABSTRACT FROM AUTHOR]

OBJECTIVES: The aims of this study were to evaluate the diagnostic accuracy of endoscopic ultrasound-guided fine needle aspiration (EUS-FNA) in patients with suspected pancreatic cancer, and to assess immediate, acute, and 30-day complications related to EUS-FNA. METHODS: All patients with suspected pancreatic cancer were prospectively evaluated. A single gastroenterologist performed all EUS-FNAs in the presence of a cytopathologist. Immediate complications were evaluated in all patients. An experienced nurse called patients 24-72 h and 30 days after the procedure. Reference standard for the classification of the final diagnosis included: surgery (n = 48), clinical or imaging follow-up (n = 63), or death from the disease (n = 47). RESULTS: A total of 158 patients (mean age 62.3 yr) under- went EUS-FNA during the study period. The mean tumor size was 32 × 26 mm. The median number of passes was three (range one to 10). Of these patients, 44% had at least one failed attempt at tissue diagnosis before EUS-FNA. The sensitivity, specificity, PPV, NPV, and accuracy of BUS- FNA in solid pancreatic masses were 84.3%, 97%, 99%, 64%, and 84%, respectively. Immediate self-limited complications occurred in 10 of the 158 EUS-FNAs (6.3%). Of 90 patients contacted at 24-72 h, 78 patients (87%) responded. Of the 90 patients, 20 (22%) reported at least one symptom, all of which were minor except in three cases (one self-limited acute pancreatitis and two emergency room visits, one of which led to admission). In all, 83 patients were contacted at 30 days, and 82% responded. No additional or continued complications were reported. CONCLUSIONS: EUS-FNA is highly accurate in identifying patients with suspected pancreatic cancer, especially when other modalities have failed. Major complications after EUS-FNA are rare, and minor complications are similar to those reported for upper endoscopy. It seems that follow-up at 1 wk might capture all of the adverse events related to EUS-FNA. [ABSTRACT FROM AUTHOR]