Noninvasive Versus Histologic Detection of Gastric Atrophy in a Hispanic Population in North America

Transcription

1 CLINICAL GASTROENTEROLOGY AND HEPATOLOGY 2006;4: Noninvasive Versus Histologic Detection of Gastric Atrophy in a Hispanic Population in North America DAVID Y. GRAHAM,* ZHANNAT Z. NURGALIEVA,* HALA M.T. EL-ZIMAITY,* ANTONE R. OPEKUN,* ARMANDO CAMPOS, LEONCIO GUERRERO, ALFREDO CHAVEZ, and VICTOR CARDENAS *Department of Medicine, Veterans Affairs Medical Center, and Baylor College of Medicine, Houston, Texas; University of Texas School of Public Health, Division of Epidemiology, El Paso Regional Campus, El Paso, Texas; and Mexican Institute for Social Security, General Hospital for Region 6, Ciudad Juarez, Chihuahua, Mexico Background & Aims: Cancer risk is directly correlated with the severity and extent of mucosal atrophy, making identification of atrophy a goal in cancer prevention programs. The aim of this study was to compare targeted histology with noninvasive testing for the identification of antral and/or corpus atrophy in North America. Methods: In a cross-sectional study of a random sample of households, 8 gastric biopsy specimens were obtained from defined locations in the antrum and corpus. Biopsies were scored for the presence of Helicobacter pylori and gastric atrophy (defined as loss of normal glandular components). Atrophy was scored by using the Sydney system and a system based on the number and location of corpus biopsies with atrophy. Patients sera were examined for pepsinogen I, pepsinogen II, and gastrin-17 (fasting and stimulated). Results: One hundred eighty volunteers, approximately 30 per age group and ranging in age from years, participated. There were 76 men. The overall weighted prevalence of a corpus atrophy was 4.7% (95% confidence interval, ). There was a significant inverse relationship between the grade of corpus atrophy and the pepsinogen I/pepsinogen II ratio (R 0.31, P <.01). We failed to confirm the usefulness of the proposed algorithm by using gastrin-17, H pylori serology, and serum pepsinogens to categorize the gastric histology. The Sydney system underestimated the prevalence of corpus atrophy by approximately 25%. Conclusion: Noninvasive testing is both possible and practical by using pepsinogen assays for the identification of the precancerous condition of moderate to severe corpus atrophy in North American Hispanic patients. Helicobacter pylori infection causes inflammation of the gastric mucosa, which might progress to gastric atrophy with intestinal metaplasia and in some patients to dysplasia (intraepithelial neoplasia) and to gastric adenocarcinoma. The incidence of gastric cancer varies in different geographic areas and is linked to the prevalence of atrophic gastritis. 1,2 Gastric cancer is a major problem in Japan, such that population-based gastric cancer prevention programs have been instituted 3 with the goal of identifying gastric cancer before it becomes incurable. The initial approach was to perform radiographic screening with barium meal to examine large segments of the population. Individuals in whom radiographic abnormalities were discovered would be referred for endoscopic examination and appropriate therapy. More recently, noninvasive techniques have been used to enrich the proportion of cases undergoing endoscopy 4 9 in whom cancer is found. Noninvasive testing is largely based on changes in serum concentrations of pepsinogens, which are the precursor proteins for the proteolytic enzymes, pepsins, produced by the stomach. The 2 major pepsinogens, pepsinogen I (PGI) and pepsinogen II (PGII), are produced in different regions of the stomach, such that measuring their respective levels can theoretically provide information about the functional integrity of the gastric antrum and corpus mucosa. PGI is produced exclusively in the oxyntic mucosa, whereas PGII is produced in the gastric antrum and corpus as well as the duodenum. A low serum PGI level and/or a low PGI level in relation to PGII level (PGI:PGII ratio) is indicative of the presence of atrophy or severe gastritis of the gastric corpus. Gastrin is produced by G cells in the gastric antrum, and its release is finely regulated by intraluminal contents, particularly the presence of protein-containing food, and by acidity. Gastrin is released in response to a rise in luminal ph and is suppressed as the ph falls below 3. The regulation of gastrin release is perturbed in the presence of an H pylori infection, resulting in mild elevations of fasting and meal-stimulated gastrin levels and an impaired response to Abbreviations used in this paper: ANOVA, analysis of variance; CI, confidence interval; EIA, enzyme immunoassay; G-17, gastrin-17; HM- CAP, high molecular weight cell-associated protein; IgG, immunoglobulin G; PGI, pepsinogen I; PGII, pepsinogen II by the American Gastroenterological Association Institute /06/$32.00 PII: /S (05)

2 March 2006 NONINVASIVE DIAGNOSIS OF GASTRIC ATROPHY ph-related regulation of gastrin release. Changes in the serum gastrin-17 (G-17) concentrations also have predictive value in relation to the functional status of the antral mucosa. For example, a low serum G-17 concentration might result from physiologic suppression of gastrin release by highly acidic stomach contents or loss of G cells as a result of moderate or severe atrophy of the antral mucosa. An exaggerated meal-stimulated gastrin is associated with 15 H pylori induced antral inflammation. Very high levels of G-17 typically reflect the physiologic response of the antral G cells to markedly reduced acid production as is found in pernicious anemia with severe corpus atrophy, achlorhydria, but normal antral mucosa. Thus theoretically, assessment of serum pepsinogens, H pylori status, and the level of G-17 should provide an overview of the functional status of the antral and corpus mucosa. For example, in pernicious anemia one expects a marked elevation in serum G-17 levels and low serum PGI levels and a low PGI:PGII ratio reflecting corpus atrophy. In contrast, a low serum PGI level and a low PGI:PGII ratio and a low G-17 would suggest the presence of antral and corpus atrophy. Recently, a panel of 4 tests (Gastropanel; Biohit Plc, Helsinki, Finland) has been introduced with the goal of providing information regarding the status of the gastric mucosa. The panel consists of assays of the levels of serum PGI, serum PGII, PGI:PGII ratio, serum G-17 levels, and H pylori immunoglobulin G (IgG) antibodies. An associated algorithm combines the results of gastrin, pepsinogen, and H pylori tests to suggest a likely diagnosis. We examined the utility of this combination of tests in a North American population with a high prevalence of H pylori infection and an increased risk of gastric cancer. Methods Study Design The study was done in Ciudad Juarez, Mexico, which is on the US-Mexico border contiguous to El Paso, Texas. This was a cross-sectional study in which volunteers were recruited from a random sample of households selected by means of a multistage cluster, self-weighted design. The survey was designed to be self-weighted and stratified by age and sex. In the first stage, census tracts were selected (ie, primary sampling unit) with probability proportional to their size. In the second, blocks were selected by using simple random sampling. All households within selected blocks were listed, and households were selected by using systematic random sampling. Finally, individuals were listed by age and invited to participate according to specific age-strata recruitment goals: 36 subjects (18 men and 18 women) in each of the following age groups: 18 29, 30 39, 40 49, 50 59, and 60 and more years of age, to oversample the older high-risk population and compare them with equal numbers of persons in younger age strata and Figure 1. The biopsy sites are shown both in a diagram and as representative endoscopic pictures. optimize the statistical testing of the study hypothesis. The sampling sites were obtained by using an updated digital city map obtained from the (Mexican) National Institute of Statistics, Geography and Informatics. Study Population The adult population consisted of both men and women age 18 years or older. Exclusion criteria included contraindications for endoscopy such as recent myocardial infarction, severe respiratory impairment, or severe cardiovascular disease. Also excluded were patients with an incomplete endoscopic examination, those with endoscopy performed for other than diagnostic reasons, patients whose endoscopic examination was done as an emergency examination as for bleeding, and patients whose treatment required immediate follow-up treatment. Those who were allergic to egg, soy, cereals, chocolate, or milk products were also excluded. All medicines that might affect gastric acid secretion, such as H 2 -receptor antagonists or proton pump inhibitors, were prohibited for at least 1 week before any part of the study. After obtaining written informed consent, a face-to-face structured interview was completed at the time of the initial appointment. The interviewer collected demographic information, as well as information regarding previous medical treatments for dyspepsia, including proton pump inhibitors and antibiotics for H pylori eradication. The study was approved by the institutional review boards at Baylor College of Medicine, The University of Texas Health Science Center, and the Mexican Institute for Social Security. Endoscopy Experienced gastroenterologists received training as well as endoscopic photographs and a map (similar Figure to 1 but in color) showing the biopsy sites. Copies were posted in the endoscopy units. Endoscopic procedures were performed at the endoscopic facility of the main hospital of the Mexican Institute for Social Security or at a private practice at the Hospital Guernika in Ciudad Juárez. Gastric mucosal biopsy

3 308 GRAHAM ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 4, No. 3 specimens were collected by using large biopsy forceps (MicroVasive Radial Jaw without needle, 2.2-mm diameter; Boston Scientific, Natick, MA). Four biopsies were collected from the antrum and 4 from the corpus Figure ( 1). Biopsies were specifically identified with respect to site, such that the results could be specifically related to the site. Site A3 was the gastric angle; B2 was taken approximately 2 4 cm proximal to the start of the first fold on the greater curve of the stomach (distal corpus). Two biopsies (1 antral and 1 corpus) were taken for culture separately, and 1 antral biopsy was taken for rapid urease testing (Hpfast; Chek-Med Systems, Camp Hill, PA). The biopsies for histology were formalin-fixed in 10% neutral formalin. The biopsies for H pylori culture were placed in labeled tubes containing transport media and immediately frozen at 70 C until shipment to the laboratories at the VA Medical Center in Houston. Histology Each specimen was placed in a separate bottle of formalin labeled as to site on the gastric map. Samples were processed at the Gastrointestinal Mucosal Pathology Laboratory in Houston, Texas, where they were embedded in paraffin wax and cut at 4 m, and serial sections were stained with the Genta triple strain. Each specimen was scored by using a visual analog scale from 0 (absent/normal) to 5 (maximal intensity) 16 for H pylori, active inflammation, and intestinal metaplasia. In addition, the type of epithelium (antral, oxyntic, transitional) was also recorded. Atrophy was defined as the loss of normal glands with and without its replacement with fibrosis, intestinal metaplasia, and/or pseudopyloric metaplasia (also called pyloric metaplasia or mucus metaplasia). There is considerable interobserver variability in scoring antral atrophy, and to reduce that variability, areas with lymphoid follicles 17 were excluded from the analysis for atrophy. The presence of pseudopyloric metaplasia was defined as mucosa that was phenotypically antral but was anatomically obtained from a region in which corpus mucosa would be expected. In addition, in all cases the final arbitrator of antral versus pseudopyloric was made on the basis of the presence of glands that stained positively for PGI, a marker for corpus 18 epithelium as previously described. Scoring of Atrophy Corpus atrophy was scored 2 ways, according to the recommendations of the updated Sydney system with the 5 biopsies recommended (A1, A3, A4, B4, and B6) Figure ( 1) and by a system based on the number and location of corpus biopsies with atrophy (the Houston system) and designed to 19 record the extent of pyloro-corporal advancement of atrophy. Because the sites recommended for biopsy with the Sydney system would theoretically systematically not sample atrophic mucosa that had not yet extended sufficiently to involve the recommended proximal sites (ie, B4 and B6; Figure 1), 19 we added a site (B5) between the gastric angle and the Sydney system proximal site (B4) and one (B2) between the antrum and the Sydney system proximal site (B6) Figure ( 1). The B2 site was approximately 4 cm proximal to the end of the corpus folds. The B5 site was intended to be located at least 3 4 cm proximal to the gastric angle and thus to provide a site between the site recommended by the Sydney system and the antrum. Unfortunately, the majority ( 90%) of the specimens were taken close to the angle and were anatomically antral. That site was therefore excluded from all analyses. The Houston system was based on the fact that the degree of damage is initially more severe in the non acid-secreting gastric antrum, but over time the damage progresses into the gastric corpus. This can be visualized as an advancing atrophic border that involves the lesser curvature more rapidly than the greater curvature As parietal and chief cells in the corpus are replaced by mucus cells, the mucosal biopsy appears phenotypically like antral mucosa and is called mucus metaplasia 17,18 or pseudopyloric metaplasia. The Houston system there - fore scored the stage of atrophy on a 4-point scale ranging from 0 none to 3 severe, with 0 atrophy at no corpus site, 1 mild or atrophy only at the lesser curve site (B4), 2 moderate or atrophy at the distal corpus site (B2) and involvement of the lesser curve site B4, and 3 severe atrophy and included the proximal greater curve site (B6) Figure (see 1 for sites). As predicted by the concept of atrophy occurring as an advancing front, all with involvement of the proximal greater curve site, B6, also had involvement of the distal site, B2. Pepsinogen and Gastrin-17 Blood Collection A fasting sample (6 8 ml) of venous blood was collected without preservatives into plastic tubes (Venoject VP- 100 SP; Terumo Somerset, NJ), which was placed into ice bath for coagulation. A food-stimulated blood sample was taken 20 minutes after a protein drink that contained 10 g of protein (Biohit Plc) was given with 100 ml of water according to the manufacturer s directions. Fasting and meal-stimulated sera were stored at 70 C until assay in Houston by using the kits provided by the manufacturer (Biohit). Evaluation of anti H pylori IgG antibodies was done by using the H pylori IgG enzyme immunoassay (EIA) test kit (Biohit Inc) and a kit for IgG antibody to the high molecular weight cell-associated proteins (HM-CAPs) of H pylori (SCANLISA H pylori HM-CAP IgG Assay; Scimedx Corporation, Denville, NJ). The results were interpreted according to the manufacturer s instructions. H pylori infection was defined as a positive EIA result. The H pylori antibodies were expressed according to the formulas included in the test kits. Levels of serum G-17, PGI, and PGII were determined by using specific EIA tests (Gastrin-17 EIA Test Kit, Pepsinogen-I EIA Test Kit, and Pepsinogen-II EIA Test Kit; Biohit). Tests were performed on a microwell plate according to the manufacturer s instructions. All EIA techniques were based on measuring the absorbance after a peroxidation reaction at 450 nm. Between the reaction steps, the plates were washed by using ELx405 Automated Plate Washer (BIO-TEK Instruments Inc, Winooski, VT). The absorbance was measured by

4 March 2006 NONINVASIVE DIAGNOSIS OF GASTRIC ATROPHY 309 using a microwell plate reader Microplate Spectrophotometer Spectra Max 190 (Molecular Devices, Sunnyvale, CA). For determination of serum PGI and G-17 values, second-order fits on standard concentrations were used to interpolate/extrapolate unknown sample concentrations automatically with the help of the BP800 built-in software. Helicobacter pylori Status H pylori infection was defined as a positive culture for H pylori or a positive histology and rapid urease tests. A presumed positive was defined as culture negative but positive histology with a score of 2 or greater in at least 2 separate biopsies (n 2). Statistical Analyses The goal was to test the performance of the assays in classifying corpus atrophic gastritis in approximately 36 individuals in each of the age strata. At the same time, the study allowed us to describe the prevalence of atrophic gastritis according to age, gender, and other risk factors including H pylori infection. Prevalence estimates for the study population used weighted analyses using SAS-callable SUDAAN (version 8.1; Research Triangle Institute, Research Triangle Park, NC). Sampling weights, ie, the number of individuals in the population that each study subject represents, were calculated by using the probability of selection of each individual and adjusting for nonresponse. Sampling weights were used only in this report to estimate overall prevalence figures. Analyses were done with Stata Software (Stata Release 8; Stata Corporation, College Station, TX). The unpaired Student t tests were used to compare the means of PGI:PGII ratios and grade of corpus atrophy. Fisher exact probability test was used from nonparametric tests to test 2 independent samples when small samples were compared. The relation between pepsinogen values and atrophy scores was evaluated by using 1-way analysis of variance (ANOVA) and Kruskal-Wallis 1-way ANOVA on ranks. The classification of the study participants according to the status of their gastric mucosa by histology (ie, gold standard) was also compared with the classification of atrophy on the basis of the tests in the Biohit panel. To that end, data were cross-tabulated, and 2 or Fisher exact tests were used to compare the proportions classified by the different tests. Sensitivity, specificity, and predictive value of a positive test result were evaluated by using standard definitions, and sensitivities were plotted against 1-specificities of the test (ie, receiveroperator characteristic curves), and the area under the curve was computed by using STATA. Results Study Population Demographics and Prevalence of Helicobacter pylori Infection A total of 180 adult volunteers ranging in age from years participated, 76 men and 104 women. The overall seropositivity rate of H pylori infection as Figure 2. The age-weighted increase in gastric antral and corpus atrophy is shown. defined above was 76%, increasing from 55.2% in the age group to 90.3% at ages There was no significant difference in the overall prevalence of H pylori infection between men and women (78.4% vs 74.5%). Prevalence of Gastric Atrophy Adequate biopsy specimens to assess corpus atrophy were available from 178 subjects, and data for antral atrophy were available on 179 subjects. Corpus atrophy was found in 16 (9%) of the study subjects. The prevalence of a corpus atrophy weighted for oversampling of older individuals was 4.7% (95% confidence interval [CI], ), (n 178); 62% of the patients were women. The mean age for women with corpus atrophy was years, and for men it was years (P.06). With unweighted analyses, no differences were found in the prevalence of corpus atrophy by gender: prevalence ratio, 1.1; 95% CI, However, corpus atrophy increased significantly with age (Figure 2), from none among those younger than 30 years of age to 2.6%, 11.6%, 9.7%, and 19.4% for those in their 30s, 40s, 50s, and 60s or more, respectively ( 2 test for linear trend, P.004). The overall unweighted prevalence of corpus atrophy among those with active H pylori infections was 9.6% (n 135) compared with 6.9% among those without an active H pylori infection (n 43). However, the CI of this difference included the null value (prevalence ratio, 1.4; 95% CI, ). Antral atrophy was found in 47 (26.3%) of the study subjects (n 179). The overall prevalence of antral atrophy was 19.6% (95% CI, ). Unweighted analyses showed that antral atrophy increased from 13.8% in the age group years and reached 52.6% in the oldest age group. No differences by gender were found, but antral atrophy was present in 30.9% of those

5 310 GRAHAM ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 4, No. 3 Table 1. H pylori Infection Serology as Detected by HM-CAP EIA and Hp IgG EIA According to Biopsy-Based Diagnosis in Residents of Ciudad Juarez, Mexico, 2004 Positive Hp status Negative HM-CAP Positive 99 (93.4%) 1 Negative 7 38 (97.4%) Biohit Positive 84 (79.2%) 1 Negative (97.4%) Abbreviation: Hp, H pylori. with active H pylori infections (n 136) compared with 11.6% (n 43) of those without active H pylori infection (prevalence ratio, 2.6; 95% CI, ). Comparison of Sydney and Houston Grading Systems for Atrophy The intention was to take a gastric biopsy at site B5, which was at least 3 4 cm proximal to the gastric angle, and thus provide a site between the site recommended by the Sydney system and the antrum. Unfortunately, the majority of the specimens were taken close to the angle and were anatomically antral, as shown by the absence of staining for PGI, the marker for corpus mucosa. This site was therefore excluded from the calculations, thus basing the comparison of the Houston system and the Sydney system on the outcome by using an additional site on the greater curve that is more distal (B2) than the Sydney site curve site (B6). The Sydney system underestimated prevalence of corpus atrophy by approximately 25% compared with the Houston system (ie, 12.3% vs 16.2% for the Sydney and Houston systems, respectively). The results were similar (24%) when the analyses were restricted to those with active H pylori infections (14.1% vs 18.5% for the Sydney and Houston systems, respectively). Helicobacter pylori Serology Sixteen subjects were excluded from the analysis of the seroprevalence of H pylori infection with the Biohit kit because a serum sample was not collected for 1 subject, and 15 had indeterminate results. Twenty-one were dropped from the analysis of the seroprevalence of H pylori infection with H pylori HM-CAP IgG assay. A serum sample was not collected from 1 subject, and 20 had indeterminate results. The overall seroprevalence of H pylori infection was comparable by using either the HM-CAP or the H pylori IgG EIA test kit (74.8%, n 159, and 74.4%, n 164, respectively) (Table 1). The anti H pylori ELISAs were evaluated against the gold standard for H pylori positivity as defined above. The sensitivity and specificity for the HM-CAP IgG test were 93% and 97% as compared with the gold standard, whereas these figures for the Biohit EIA were 79% and 97%, respectively. Thus the Biohit kit had a larger yield of false-negative results Table ( 1). Serum Pepsinogen I and II and Helicobacter pylori Infection Serum samples from 179 subjects were available for testing. The serum PGI levels in H pylori infected individuals were higher than among those without an active H pylori infection ( vs , respectively) (P.01). Similarly, the serum PGII levels were higher in those with H pylori infection ( vs 6.7 3, respectively) (P.01). Serum Pepsinogen I and II and Corpus Atrophy There was an inverse relationship between the PGI:PGII ratio and the grade of corpus atrophy (Figure 3) (P.01). The comparable values for sensitivity and specificity of 75% and 71.2% for this population were achieved by using receiver operating conditions analysis with a cutoff point of 6.7. When tested against a cutoff of 3.2, the sensitivity and specificity were 25% and 95.7%, respectively. Gastrin-17 and Atrophy G-17 levels are increased after ingestion of food such that a low fasting and postprandial serum G-17 level is a potential marker for the presence of antral atrophy. 25 Seventy-three subjects had a fasting and postprandial G-17 level below 25 g/l (mean, Figure 3. Box plots showing the inverse relationship between the severity of corpus atrophy and the ratio of serum (S)-PGI to PGII.

6 March 2006 NONINVASIVE DIAGNOSIS OF GASTRIC ATROPHY 311 Figure 4. GastroPanel algorithm for detection of antral and corpus atrophy among those with negative anti H pylori (Hp) IgG. fasting increasing to g/l) (P.001). Antral atrophy was present in 20, with 14 having grade 1, 1 with grade 2, 3 with grade 3, and 1 each with grades 4 and 5. The mean fasting and postprandial G-17 levels were similar for those with low gastrins and no antral atrophy (score 0) and those with scores of 3 or greater (eg, vs and vs g/l for fasting and postprandial G-17 levels for those without and with atrophy, respectively). Five subjects with antral atrophy with grades of 3 or greater had fasting, postprandial, or both levels greater than 25 g/l. The sensitivity and specificity for low fasting and postprandial G-17 levels were 50% and 58.5%, respectively. Ten subjects had fasting G-17 levels greater than 60 g/l, and 4 had levels greater than 100. Three of the 4 had corpus atrophy (grade 1 once and grade 3 in two), and all 4 had PGI:PGII ratios of 3.6 or less. However, only 50% of those with grade 2 or 3 (moderate or severe) corpus atrophy had fasting G-17 levels greater than 60 g/l. Evaluation of the Biohit Algorithm for Clinical Diagnosis The GastroPanel suggests an algorithm that uses the result of the H pylori serologic testing, G-17 levels, and serum PGI levels to provide a likely diagnosis of the status of atrophy of the antral and corpus mucosa (Figures 4 and 5 ). Fasting G-17 values and H pylori serology were available for 160 subjects. Postprandial gastrin samples were available for 163 subjects. The outcome of using the GastroPanel algorithm compared with the gold standard of histology are shown in Tables 2 and 3. The best outcome was the ability of the algorithm to correctly characterize absence of atrophy by using the postprandial serum in the group with negative H pylori serology with a sensitivity of 100% but a specificity of only 12%. The sensitivity with fasting sera was essentially the same (95%). In those with positive H pylori serology the postprandial sera again showed the best results, with a sensitivity of 88% but again with a specificity of 27%. In all other comparisons for both H pylori serology positive and negative cases, the sensitivity was less than 45%. Discussion It is now generally accepted that epidemic noncardiac gastric cancer is etiologically associated with H pylori infection, and that the presence of atrophic gastritis 1 is the primary easily identifiable risk factor. Changes in environmental factors including diet, food storage, and sanitation have resulted in a major reduction in the prevalence of H pylori and of gastric cancer among those 26 with the infection. Eradication of H pylori at a nonatrophic stage will likely prevent the subsequent development of gastric cancer. In contrast, eradication among those with advanced atrophic changes will likely reduce the incidence but not eliminate the risk of the subsequent development of cancer. Thus, it has been sug - 27 gested that the combination of H pylori eradication and posteradication surveillance for those with atrophic gastritis might be the best strategy. 27 Noninvasive testing for detection of the presence of atrophic changes in the antrum and corpus would be the most efficient and cost-effective method of selecting patients for endoscopic evaluation for primary prevention and for identifying those who might benefit from longer-term surveillance after H pylori eradication. 27 The majority of large and proof-of-principle studies have been done in Japan. Recently, the concept 28 of using a panel of tests has been proposed and has been tested in Russia, Scandinavia, and Italy The Figure 5. GastroPanel algorithm for detection of antral and corpus atrophy among those with positive anti H. pylori (Hp) IgG.

7 312 GRAHAM ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 4, No. 3 Table 2. GastroPanel Algorithm Using Sera That Were H pylori Positive or H pylori Negative (Figures 4 and 5 for Details) Histology Panel N TP TN FP FN Se Sp Diagnosis Hp negative No atrophy (fasting) (fed) Atrophic gastritis, corpus or antrum, autoimmune or Hp (fasting) (fed) Corpus atrophic gastritis autoimmune (fasting) (fed) Diagnosis: Hp positive Atrophic gastritis, corpus or antrum, Hp (fasting) (fed) Antrum atrophy (fasting) (fed) Atrophic gastritis, corpus or antrum, autoimmune or Hp (fasting) (fed) Non atrophic gastritis (Hp ) (fasting) (fed) Atrophic gastritis corpus (Hp ) (fasting) (fed) Non atrophic gastritis (Hp ) (fasting) (fed) Abbreviations: TP, true positive; TN, true negative; FP, false positive; FN, false negative; Se, sensitivity; Sp, specificity; Hp, H pylori. Hispanic population in North America is a high-risk group for development of gastric cancer, with gastric cancer being the second leading cause of cancer deaths in Mexico in 2003 ( htdocs/estadisticas/mortalidad/mortalidad.htm, accessed July 22, 2005). Our study confirmed that a low ratio of serum PGI: serum PGII was indicative of moderate or severe corpus atrophy, and a ratio below 6.5 was able to accurately diagnose 76% (95% CI, 69% 82%) of patients with atrophic corpus gastritis Figure ( 6). Larger scale studies are needed to better refine the best cutoff values. The place for G-17 testing in relation to evaluating the status of the gastric mucosa remains unclear. G-17 levels are increased in H pylori infections but typically remain within the normal range unless antral atrophy has occurred. 15 A high G-17 level implies low acid secretion, which in the absence of anti-secretory drug therapy implies corpus atrophy. The presence of a low fasting and a normal postprandial G-17 level is thought to reflect acid-associated suppression of gastrin release. In contrast, a low fasting and postprandial gastrin level is postulated to reflect antral atrophy; however, in this study it had a relatively low sensitivity and specificity for antral atrophy. High fasting G-17 levels were restricted to those with severe corpus atrophy. The presence of a low pepsinogen and high G-17 level has been suggested as a marker for those with severe atrophic gastritis and a high risk of having 11,33 or developing gastric cancer. This hypothesis has recently been confirmed in a study in which the combination of serum pepsinogen testing (eg, PGI 45 ng/ml and serum G pg/ml) gave the best sensitivity and specificity for identifying cancer 9 among high-risk patients. We tested the proposed GastroPanel algorithm to evaluate whether it would provide an accurate estimate of the presence or absence of antral or corpus Table 3. GastroPanel Algorithm Using All Sera With the Outcome Shown in Figure 4 Diagnosis: Hp positive or negative Histology Panel N TP TN FP FN Se Sp No atrophy (fasting) (fed) Atrophic gastritis, corpus or antrum, autoimmune or Hp (fasting) (fed) Corpus atrophic gastritis autoimmune (fasting) (fed) Abbreviations: Hp, H pylori; TP, true positive; TN, true negative; FP, false positive; FN, false negative; Se, sensitivity; Sp, specificity.

8 March 2006 NONINVASIVE DIAGNOSIS OF GASTRIC ATROPHY 313 Figure 6. The accuracy of the PGI:PGII ratio or detecting corpus atrophy is shown in relation to different cutoff values. The area under the receiver-operator characteristic curve was (95% CI, ). atrophy in unselected individuals. The results were disappointing, suggesting that, at least in this population and similar to those in another asymptomatic population from Italy, 31 the algorithm might need to be revised for use in low-risk populations. The current algorithm starts with H pylori status, and although it proved useful to identify normal stomachs, it did poorly for identifying atrophic gastritis. It is possible that starting with the PGI levels or PGI:PGII ratio might provide more useful information. Recently it has been reported that high ghrelin levels correlate 34 with low pepsinogen levels and atrophic gastritis. It is unclear whether measurement of ghrelin levels would enhance the noninvasive detection of atrophic gastritis, and more studies are needed. References 1. Graham DY. Helicobacter pylori infection is the primary cause of gastric cancer. J Gastroenterol 2000;35(Suppl 12): Sipponen P, Kekki M, Haapakoski J, et al. Gastric cancer risk in chronic atrophic gastritis: statistical calculations of cross-sectional data. Int J Cancer 1985;35: Oshima A. Secondary prevention: screening methods in highincidence areas. In: Sugimura T, Sasako M, eds. Gastric cancer. Oxford: Oxford University Press, 1997: Aoki K, Misumi J, Kimura T, et al. Evaluation of cutoff levels for screening of gastric cancer using serum pepsinogens and distributions of levels of serum pepsinogen I, II and of PG I/PG II ratios in a gastric cancer case-control study. J Epidemiol 1997;7: Fukuda H, Saito D, Hayashi S, et al. Helicobacter pylori infection, serum pepsinogen level and gastric cancer: a case-control study in Japan. Jpn J Cancer Res 1995;86: Hattori Y, Tashiro H, Kawamoto T, et al. Sensitivity and specificity of mass screening for gastric cancer using the measurement of serum pepsinogens. Jpn J Cancer Res 1995;86: Kitahara F, Kobayashi K, Sato T, et al. Accuracy of screening for gastric cancer using serum pepsinogen concentrations. Gut 1999;44: Miki K, Ichinose M, Ishikawa KB, et al. Clinical application of serum pepsinogen I and II levels for mass screening to detect gastric cancer. Jpn J Cancer Res 1993;84: Shiotani A, Iishi H, Uedo N, et al. Histologic and serum risk markers for noncardia early gastric cancer. Int J Cancer 2005; 115: Miki K, Ichinose M, Shimizu A, et al. Serum pepsinogens as a screening test of extensive chronic gastritis. Gastroenterol Jpn 1987;22: Kekki M, Samloff IM, Varis K, et al. Serum pepsinogen I and serum gastrin in the screening of severe atrophic corpus gastritis. Scand J Gastroenterol Suppl 1991;186: Varis K, Kekki M, Harkonen M, et al. Serum pepsinogen I and serum gastrin in the screening of atrophic pangastritis with high risk of gastric cancer. Scand J Gastroenterol Suppl 1991;186: Graham DY. Helicobacter pylori and perturbations in acid secretion: the end of the beginning. Gastroenterology 1996;110: Graham DY, Dore MP. Perturbations in gastric physiology in Helicobacter pylori duodenal ulcer: are they all epiphenomena? Helicobacter 1997;2(Suppl 1):S44 S Graham DY, McGuigan JE. Interactions between gastrin, gastric secretion, and Helicobacter pylori in health and disease. In: Tache Y, Goto Y, Ohning G, Yamada T, eds. Gut-brain peptides in the new millenium. Los Angeles: Cure Foundation, 2002: El-Zimaity HM, Graham DY, Al-Assi MT, et al. Interobserver variation in the histopathological assessment of Helicobacter pylori gastritis. Hum Pathol 1996;27: Rugge M, Correa P, Dixon MF, et al. Gastric mucosal atrophy: interobserver consistency using new criteria for classification and grading. Aliment Pharmacol Ther 2002;16: El-Zimaity HM, Ota H, Graham DY, et al. Patterns of gastric atrophy in intestinal type gastric carcinoma. Cancer 2002;94: Ricuarte O, Gutierrez O, Cardona H, et al. Atrophic gastritis in young children and adolescents. J Clin Pathol 2005;58: Kimura K. Chronological transition of the fundic-pyloric border determined by stepwise biopsy of the lesser and greater curvatures of the stomach. Gastroenterology 1972;63: Kimura K, Takemoto T. An endoscopic recognition of the atrophic border and its significance in chronic gastritis. Endoscopy 1969; 1: Miederer SE, Paul F, Stadelmann O, et al. Pyloric-kardiate expansion der gastritis. Endoscopy 1969;1: Kimura K, Satoh K, Ido K, et al. Gastritis in the Japanese stomach. Scand J Gastroenterol Suppl 1996;214: Satoh K, Kimura K, Taniguchi Y, et al. Distribution of inflammation and atrophy in the stomach of Helicobacter pylori-positive and -negative patients with chronic gastritis. Am J Gastroenterol 1996;91: Sipponen P, Ranta P, Helske T, et al. Serum levels of amidated gastrin-17 and pepsinogen I in atrophic gastritis: an observational case-control study. Scand J Gastroenterol 2002;37: Graham DY. Helicobacter pylori infection in the pathogenesis of duodenal ulcer and gastric cancer: a model. Gastroenterology 1997;113: Graham DY, Shiotani A. The time to eradicate gastric cancer is now. Gut 2005;54: Sipponen P, Harkonen M, Alanko A, et al. Diagnosis of atrophic gastritis from a serum sample. Clin Lab 2002;48: Hartleb M, Wandzel P, Waluga M, et al. Non-endoscopic diagno-

9 314 GRAHAM ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 4, No. 3 sis of multifocal atrophic gastritis; efficacy of serum gastrin-17, pepsinogens and Helicobacter pylori antibodies. Acta Gastroenterol Belg 2004;67: Pasechnikov VD, Chukov SZ, Kotelevets SM, et al. Invasive and non-invasive diagnosis of Helicobacter pylori-associated atrophic gastritis: a comparative study. Scand J Gastroenterol 2005;40: Ricci C, Vakil N, Rugge M, et al. Serological markers for gastric atrophy in asymptomatic patients infected with Helicobacter pylori. Am J Gastroenterol 2004;99: Vaananen H, Vauhkonen M, Helske T, et al. Non-endoscopic diagnosis of atrophic gastritis with a blood test: correlation between gastric histology and serum levels of gastrin-17 and pepsinogen I: a multicentre study. Eur J Gastroenterol Hepatol 2003; 15: Hallissey MT, Dunn JA, Fielding JW. Evaluation of pepsinogen A and gastrin-17 as markers of gastric cancer and high-risk pathologic conditions. Scand J Gastroenterol 1994;29: Suzuki H, Masaoka T, Hosoda H, et al. Plasma ghrelin concentration correlates with the levels of serum pepsinogen I and pepsinogen I/II ratio: a possible novel and non-invasive marker for gastric atrophy. Hepatogastroenterology 2004;51: Address requests for reprints to: Victor Cardenas, MD, MPH, PhD, UT School of Public Health El Paso Regional Campus, 1100 N Stanton Avenue, Suite 110F, El Paso, Texas victor. cardenas@uth.tmc.edu; fax: (915) Supported in part by the Office of Research and Development, Medical Research Service, Department of Veterans Affairs, by Public Health Service grant DK56338 which funds the Texas Gulf Coast Digestive Diseases Center, and by a grant from Biohit Diagnostics, Helsinki, Finland. No for-profit company had any input into the study design or analyses. Dr Graham receives royalties from Baylor College of Medicine from a patent on the antigen used in the HM-CAP serologic test. The Mexican Institute for Social Security authorized the use of the General Hospital Region 6 endoscopic facility and staff support to conduct the endoscopies.