current infection and is not useful among populations with natural loss of the H. pylori infection (e.g., gastric cancer patients). The data suggest that among dyspeptic patients, the absence of 37 K antigen and the presence of 35 K antigen might be a useful markers for increased risk of gastric cancer and gastric ulcer, respectively.

S1236

Do Serum Pepsinogen and Gastrin Predict Gastric Intestinal Metaplasia? Chiara Ricci, Massimo Rugge, Nimish Vakil, Luigi Gatta, Federico Perna, Valentina Rnsso, Gioacchino Leandro, Marcello Menegatti, Mario Miglioli, Dmo Vaira

Studtes from Japan have suggested that serum tests may be an alternative to endoscopy in the diagnosis of intestinal metaplasia. We assessed whether serum gastrin (G) or serum pepsinogen I (Pgl) were accurate tools to diagnose the presence of atrophy and intestinal metaplasia (AIM) in the gastric antrum or corpus in healthy asymptomatic subjects with or without Helicobacter pylori (HP) infection. Methods: 147 asymptomatic subjects (60 male; mean age 56.52 yrs phis/minns 10.90 SD) underwem endoscopy with histology, rapid urease test and culture. A blood sample was also obtained after an overnight fasting to assess gasttin (gastrin opensquare 125 closesquare Radioimmunoassay kit) and Pg I (Eiken Chemical CO LTD, Tokyo, Japan). The expected normal range of G and PgI in healthy volunteers is 20- 100 pg/ml and 25-75 microg/1. Intestinal metaplasia was scored according to the Updated Sydney System and was confirmed by High Iron Diamine (HID) staining. Results. There were 45 patients with atrophy and intestinal metaplasia, 32 with antral IM and 13 with corpus IM. All patients with intestinal metaplasia had accompanying atrophy. Using > 100pg/ ml and <25microg/1 as cut-off for G and Pgl respectwely, Table 1 shows the sensitivity, specificity and likelihood ratios for gasmn and pepsinogen in predicting intestinal metaplasia. Conclusions: Serum Gastrin/pepsinogen are specific but not sensitive for the diagnosis of intestinal metapfasia.

Sensitivity Specificity LRwe LR-vo Antrum Ill 0%(0.0,107) 98.3% (93.9- 0 (0-6.629) 1.018 (1.014- G(~%(:.I,) 99.5) 1,065) Antrum I i P01 1,065 (1.051- (95%C.1.) 0%(0-0.110) 93.9(87.9-97) 0(0-1.867) 1,138) Coq~tm IM G 0%(00.228) 98.5% (94.7- 1.015 (1.014- (95%C,I,) 99.5) 0 (0-17,974) 1.056) Cocpoll IM P91 15.4% (4,3- 2.901 (0~671- 0.894 (0,706- (95%C,1.) 42.2) 94.7 (89.5-97.4) 12.55} 1.131)

S1237

Role of Detection of Urine H.pylori Antibody and Measurement of Serum Pepsinogen Levels in the Screening of Gastro-Duodenal Diseases Atsushi Takagi, Ryuzo Deguchi, Kermji Kobayashi, Takayuki Shirai, Tohru Endoh, Kozo Nagai, Hiroshi Takahashi, Ryouichi Suzuki, Katsunori Saigennji, Norifumi Saitoh, Wasaburo Koizumi

Helicobacter pylor~ infection is associated with peptic ulcer disease and gastric cancer. How- ever, many patients with H.pylori infection remain asymptomatic. Measurement of serum pepsinogen l/II has been proposed as a useful method of gastric cancer screening.Therefore, we investigated the usefulness of measurement of H.pylori antibody and serum pepsinogen levels as a screening test for patients with and without dyspepsia. Methods: The subjects were 387 patients ( 243 men and 144 women, mean : 55.7ys) scheduled for upper endoscopy at 7 hospitals in Kanagawa prefecture. Informed consent was obtained from all patients and the protocol was approved by the ethics committee of each institution. Urine antibody to H.pylo~ was measured by ELISA(urinelisa, Otsuka, Japan). Levels of serum pepsinogen I (PGl)and pepsinogenll(PGII) were determined by radloimmunoassay. PGI/II ratio under 3.0 and PG I < 70 ng/ml was considered indicative of atrophic gastritis. H.pylon status was determined using urea breath test, rapid urease test, culture or histological examination. Histological gastritis was evaluated by updated Sydney system. Results: Two-hundred five Patients had dyspeptic symptom, and 182 patients were asymptomatic. The sensitivity and specificity of urine antibody to Hpylori were 97.9% and 58.7%, respectively. The prevalence of H.pylori in patients with dyspepsia and without symptom was 83.5% and 74.6%, respec- tively. The prevalence of H.pylor~ in patients with GU, DU, gastric cancer, NUD was 99/ 108 (91.7%), 66/80 (82.5%), 45/51 (88.2%), and 43/74 (58.1%), respectively. There was no difference in detection rate in peptic ulcer patients with and without dyspeptic symptom. Furthermore, 12 out of 51 patients with gastric cancer were asymptomatic. The sensitivity and specificity of PG measurement to gastric atrophy were 38.4%, and 97%, respectively. There was no increase in the number of GU and DU patients using a combination of H.pylori antibody with PG measurement. One more patient with gastric cancer was detected using the combination of H.pylori antibody and PG measurement. Conclusion: Detection of H.pylori infection using urine H.pylori antibody is a useful method not only for dyspeptic patients but also for asymptomauc patients. Although PG measurement is highly specific in detecting atrophic gastritis, it appear to be less significant in combination with H.pylori antibody testing for detecting gastro-duodenal diseases including gastric cancer.

51238

Serum Pepsinogen II for Early Assessment of Success of H. Pylori Eradication kucas Cavallaro, All M. Moussa, Roberta Merli, Veronica Iori, Giulia M. Cavestro, Elena Cerati, Nadia Ahavilla, Nadia Dal Bo', Massimo Rugge, Giancarlo Colla, Anna Bertele', Ange[o Franae', Francesco Di Mario

Background: Hehcobacter pylon stool antigen (HpSA) is a useful non invasive tool for diagnosis and for check after eradication therapy of Hp infection. Serum pepsinogen II (sPGI[) levels were found increased in H. Pylon related gastritis. Aim: to evaluate the relationship between sPGII levels and HpSA (at first diagnosis of the infection and in post

eradication control) Patients and Methods: A total of sixty four consecutives patients with dyspeptic symptoms were studied. 40 patients ( 23 F; mean age 51, range I9-82 years ) with unknown Hp status ( Group A) were tested for Hp by means HpSA in first diagnosis ( Meridian, Milan, Italy) and 24 patients ( 12 F; mean age 54, range 20-80 years ) after eradication therapy ( Group B). HpSA was used to assess the efficacy of eradication therapy. In all patients, serum sample was collected for the determination of SPGII levels ( EIA, Biohit Helsinka Finland) Normal value of sPGII levels is considered 2-10 mcg/L The comparison of groups was made using nonparametric statistics (Mann-Whitney U Test). Results: In group A: 21 patients resulted Hp-ve with mean sPGII levels 7.5 mcg/L (SD 3.1) and 19 patients Hp+ve showed mean sPGII levels of 14.2 mcg/L (SD 62); a statistically significant difference was found between the two groups according the Hp status ( p = 0001). In group B: 20 patients Hp-ve, presented mean sPGII levels 9.0 mcg/L (SD 5) and 4 patients resulted not eradicated for Hp, presented mean sPGII 15 mcg/L (SD 10) : there are no statistically significant difference between the Hp + ve and Hp-ve patients. Conclusions: serum pepsinogen II could be proposed as non invasive test in first diagnosis or in post eradi- cation.

81239

The Influence of CAG a Positivity on the Accuracy of Non Invasive Diagnostic Methods and Eradication Rate in Helicobacter Pylori Infection Istvan Racz, Emese Babarczy, Andrea Szabo, Gyula Pecsi, Mihaly Csondes, Katafin Bircher, Artur Nemeth

Background: The effect of pathogeneticy marker Cag A on the sensitivity of diagnostic tests in Helicobacter pylon (Hp) infection is unknown. It is also poorly investigated yet, whether the Cag A status influence the eradication effect in Hp positive patients. A~ms: In our prospective study we investigated the effect of cag A status on the efficacy of the different non invasive diagnostic methods in Hp positive duodenal ulcer (DU) and ulcer predominant functional dyspeptic (FD) patients. We also tested whether Cag A positivity influenced the effect of Hp eradication therapy. Patients and methods: 20 Hp positive, DU and 10 Hp positive FD patients were included. At inclusion Hp infection was tested by RUT, UBT and histology paralelly; Hp infection was accepted in positivity of at least two of these tests. Eradication was performed with one week RBC + clarithromycin + amoxicillin regimen followed by 4 week ranitidin therapy, which was stopped one week before the final visit. At the 6 week control all diagnostic methods were repeated while serum anti Hp IgG antibodies were measured quantitatively at inclusion and at the end. Results: Cag A positivity was detected in 11 (55%) DU patients, and in 3 (30%) FD patients.The UBT DOB values were similar in the Cag A negative and positive cases (12.5 % vs. 12.6%) The mean values of anti Hp IgG antibodies in Cag A positive and negative patients did not differ significantly (42+-12 U/ml vs. 21+-3,5 U/ml; p<0.I) . Similarly, no significant different was found between the means of anti Hp IgG levels of DU and FD patients (35 + -9 U/ml vs 22 + -5 U/ml; p>0 .1) Two out of three patients with eradication failure were Cag A positive. Conclusions: 1. Cag A positivity had no significant influence on the quantitative measures of non invasive Hp diagnostic tests 2. We observed a link between eradication failure and Cag A positivity.

$1240

Disappearance of Gastric Hyperplastic Polyps after eradication of Helicobacter pylori: CompaMson between Responder and Non-Responder Cases Toshifumi Ohkusa, Hiroto Miwa, Mariko Hojo, Jiro Kumagai, Tom Tanizawa, Daisuke Asaoka, Takeshi Terai, Ryuichi Ohknra, Nobuhiro Sato

Background: Several studies, including our previous work, have indicated that eradication of H. pylori leads to the disappearance of hyperplastlc polyps m the stomach However, there are a few reports of non-disappearance cases after eradication. Aim: Our aim was to compare endoscopic and serologic findings of responder and non-responder cases of hyperplastic polyps to try to determine the cause(s), other than H. priori infection, of the formation or growth of gastric hyperplastic polyps. Methods: We retrospectively studied 33 patients whose hyperplastic polyps disappeared after eradication of H. pyIori, and 10 patients whose hyperplastic polyps did not disappear following eradication. These patients were endoscopically and serologically examined before, 1-3 months after, and 12-15 months after H. pylor~ eradication. Results: The Responder and Non-responder groups were similar with respect to age, sex, coexisting diseases, and histologic findings. The number and maximum size of polyps tended to be larger in the Non-responder group than in the Responder group before treatment. The serum gastrin level was higher in the Non-responder group than in the Responder group before, 1-3 months after and 12-15 months after the eradication (p=0.0096, p>0.2, p=0.0014). Serum pepsinogen I1 level was also higher in the Non- responder group than in the Responder group at 12-15 months post H. pylori eradication (p = 0.037). On histologic examination, similar reductions in the degree of inflammatory cell infiltration in the gastric mucosa of the antrum and body were seen in both the Responder and Non-responder groups. Conclusions: The persistent high gastrin level as well as high pepsinogen II level may sustain hyperplastic polyps of the stomach

$1241

Prospective Study to Determine if Endoscopic Status of Gastrit is Is Associated with Specific Helicobacter Pylori Genotypes Containing Virulent Genes Ruba A. Abdelhadi, Raymond Podzmski, Diane Podzorski, Raja Rabah, Ronald Thomas, Vasundhara Tolia

Our aim was to evaluate genotypes as virulence factors in determining Helicobacter pylori gastritis severity on endoscopy. All children underwent endoscopy for the usual indications, and gastritis was graded as nodular or non-nodular. Gastric biopsies from patients with H. pylori gastritis were analyzed for the presence of the vacA, cagA, iceA, cagE (also called picB), and babA2 genes as virulence genes using PCR DNA from clinical isolates was extracted using IsoQuick procedure, with subsequent amplification of specific virulent genes.

A-179 AGA Abstracts