Abstracts of the 19th National Congress of Digestive Diseases / Digestive and Liver Disease 45S (2013) S55–S218
hospitals; the patients ﬁlled out to questionnaire before starting colonoscopy. Bowel cleansing were assessed by Ottawa scale; colonoscopy was assessed as “effective” in case of grade A, B, C. Results: A complete colonoscopy was achieved in 100% of cases. Effectives colonoscopies in A group where 87.8%, in B group 89.1%. In A group effectivee colonoscopies with better degree of bowel cleansing (grade A and B of Ottawa scale) where 53.34%, in B group 39.06% (n.s.). Bowel-preparation intake (100% of solution) in A group was 88.89% and in B group 98.31; however most of patients took >75% of solution (97.7% in A group and 98.4% in B group). Intake difﬁculty was deﬁned “moderate” or “high” by 33.3% of patients in A group and by 35.9% of patients in B group. However in 40% of patients in A group reported mild side effects, in B group only 34.38%. Conclusions: in those patients who cannot take split-dose regimen before colonoscopy, the number of effectives colonoscopies is the same for both regimens. But in the A group the number of effectives colonoscopies with higher grade of bowel cleansing is greater: in this case medical examiner is not compelled to use aspiration and requested time to complete examination is shorter. Side effects are more frequents in A group (the difference is minimal). Intake difﬁculty is mostly the same in A and B group, but the number of speciﬁcs side effects is slightly higher in A group.
P.16.1 DIAGNOSTIC VALUE OF EUS IN THE SELECTION OF PATIENTS WITH GASTRIC CANCER ELIGIBLE FOR A NEOADJUVANT CHEMOTHERAPY F. Antonini ∗ ,1 , W. Siquini 2 , S. Piergallini 1 , V. Belﬁori 1 , B. Marraccini 1 , M. Lo Cascio 1 , C. Manfredi 1 , G. Macarri 1 1 Ospedale
A. Murri, UOC Gastroenterologia ed Endoscopia Digestiva, Fermo, Italy; 2 Ospedale Madonna del Soccorso, UOC Chirurgia, San Benedetto del Tronto, Italy Background and aim: Preoperative evaluation of gastric cancer (GC) is increasingly important to differentiate patients who will beneﬁt from immediate surgical resection (T stage 1–2, N0) from those with locally advanced GC (LAGC) who are better candidates for neoadjuvant chemotherapy (T stage 3–4, or any N+). Aim: To evaluate the accuracy of EUS in the preoperative staging of GC with regard to select patients eligiblility for neoadjuvant chemotherapy. Second aim was to identify histopatological features that can affect the accuracy of EUS. Material and methods: Between September 2011 and September 2012, 26 consecutive patients (17 male, median age 69 years, range 43–87 years) with histologically conﬁrmed GC who underwent primary resection, were studied with EUS (GF-UE160, Olympus, Hamburg, Germany) before surgery. Preoperative EUS staging (T1 to T4 and N+ or N–) was compared to the postoperative pathological ﬁndings, according to 2010 TNM classiﬁcation. Roundish, hypoechoic and well-demarcated lymph-nodes were considered as malignant. Pathological features about tumor size and location (upper, middle, lower), histological type (differentiated, undifferentiated) were considered. Results: The overall staging accuracy of EUS for T was 73% (19/26), and for T1, T2, T3 and T4 was 100% (5/5), 0% (0/1), 33% (2/6) and 87% (12/14), respectively. The N staging accuracy was 69% (18/26) overall, and 83% (10/12) for N+ and 57% (8/14) for N–. The overstaging and understaging rates was 7% and 19% for T, and 7% and 30% for N, respectively. Every EUS understaging was for pathological T4, diagnosed as T3 in 4 cases (23.5%) and as T2 in one (5.8%). The accuracy, sensitivity and speciﬁcity of EUS For LAGC was 92%, 95% and 83%, respectively. None of the histopatological factors (tumor size, location and histology) signiﬁcantly inﬂuence EUS results. Conclusions: EUS is extremely useful to select patients eligible for a neoadjuvant chemotherapy as it provides the most accurate assessment of wall inﬁltration and lymph nodes involvement. The major concern in our study is understaging in patients with T4 tumor. However only in 5.8% of cases this could alter the condition of LAGC thus determining change in treatment. In the era of neoadjuvant chemotherapy, with the intention to provide the optimal treatment for each patient, it is suggested that EUS should be always used in the preoperative staging algorithm for GC.
P.16.2 EUS-GUIDED FINE NEEDLE TATTOOING (EUS-FNT) FOR PREOPERATIVE LOCALIZATION OF SMALL PANCREATIC NEUROENDOCRINE NEOPLASMS: A SINGLE CENTER EXPERIENCE G. Vanella, S. Alﬁeri, F. Attili ∗ , D. Galasso, F. Tsiopoulos, F. Rosa, A. Fusco, G.B. Doglietto, L. De Marinis, G. Costamagna, A. Larghi Ospedale Gemelli, Roma, Italy Background and aim: Small pancreatic neuroendocrine neoplasms (PNENs) can be difﬁcult to detect intraoperatively and accurate preoperative localization can be important to decide the type of operation and the extent of the resection. To facilitate intraoperative detection, we performed preoperative endoscopic ultrasound guided ﬁne needle tattooing (EUS-FNT) of cytologically proven pancreatic NETs and evaluated the safety and efﬁcacy of this new technique. Material and methods: Eleven consecutive patients (mean age 51.9±15.6 yrs; M/F 4/7) underwent EUS-FNT of an insulinoma (7) and of a nonfunctional PNENs (4). The procedure was carried out using the standard linear EUS scope in 9 patients and the newly developed forward viewing linear echoendoscope (XGIF-UCT160J-AL5, Olympus Medical Systems, Tokyo, Japan) in 2 patients. Tattooing was performed by injection of a sterile, biocompatible, non-pyrogenic suspension containing highly puriﬁed carbon particles (Spot, GI Supply, Comp Hill, PA, US) directly into the lesion using a 22 gauge needle (Echotip Ultra, Cook Medical, Winston-Salem, N.C.). Results: The lesions were located in the pancreatic tail (4), body (3), isthmus (2), head (1), and uncinate (1). No interference or disturbance from the previously performed EUS-FNA was observed. The mean size of the lesions was 11.3±2.7 mm. Resection of pancreatic NET was performed with a median of 9 days (range 1–61 days) after the tattoo. Six patients underwent open and ﬁve laparoscopic surgeries. The tattooed lesions were easily recognized at surgery in all but one patient. In the latter patient PNEN was located in the center of the pancreatic head and its recognition at surgery was difﬁcult and required intraoperative ultrasound. In one additional patient a small hematoma and retracted capsule of the pancreas was found intraoperative. Pathological examination of the resected specimens showed that tattooing did not troubled with the anatomical evaluation of the resected pancreas and the margins of the specimen could be easily evaluated for tumor inﬁltration. Conclusions: Our results suggest that EUS-guided FNT is a safe and useful method to mark preoperatively small pancreatic PNENs to help in their recognition at surgery. Before EUS-FNT becomes a standard preoperative marking technique in this patient population, further multicenter prospective studies are warranted.
P.16.3 LEARNING CURVE FOR ENDOSCOPIC ULTRASOUND-GUIDED FINE NEEDLE ASPIRATION (EUS-FNA) IN DIAGNOSIS OF ABDOMINAL AND MEDIASTINAL MASS E. Mannisi ∗ , A. Caruso, M. Coppola, V. Formica, I. Portarena, P. Sileri, F. Pallone, G. Del Vecchio Blanco Università degli studi di Roma Tor Vergata, Roma, Italy Background and aim: Endoscopic ultrasound guided ﬁne needle aspiration (EUS-FNA) is a well established technique associated with echo endoscopy to assess lesions of gastrointestinal tract and adjacent organs. Diagnostic yield of EUS-FNA varies according to expertise of endosonographer and cytopathologist. The minimum number of EUS-FNA recommended to reach a diagnostic accuracy of 80% ranges from 30 to 50. Aim: To investigate a single operator learning curve in a consecutive series of EUS-FNA without on-site cytopathologist. Material and methods: Sensitivity, speciﬁcity and diagnostic accuracy were evaluated in 73 EUS-FNAs which were divided in group A (early 30 EUSFNA) and group B (successive EUS-FNA). We performed further analysis, sub-stratiﬁed EUS–FNAs according to site of lesion and needle size. Results: Sensitivity and diagnostic accuracy in group B were signiﬁcantly higher than in group A (83% and 88% versus 49% and 57%, respectively; p=0.005). Speciﬁcity was 100% in both groups. A similar increase in diag-