Eiji Umegaki, Takanori Kuramoto, Yuichi Kojima, Sadaharu Nouda, Kumi Ishida, Toshihisa Takeuchi, Takuya Inoue, Satoshi Tokioka and Kazuhide Higuchi

Transcription

1 ORIGINAL ARTICLE Geranylgeranylacetone, a Gastromucoprotective Drug, Protects Against NSAID-induced Esophageal, Gastroduodenal and Small Intestinal Mucosal Injury in Healthy Subjects: A Prospective Randomized Study Involving a Comparison with Famotidine Eiji Umegaki, Takanori Kuramoto, Yuichi Kojima, Sadaharu Nouda, Kumi Ishida, Toshihisa Takeuchi, Takuya Inoue, Satoshi Tokioka and Kazuhide Higuchi Abstract Objective A treatment strategy to inhibit nonsteroidal anti-inflammatory drug (NSAID)-induced small intestinal lesions has not yet been established. To clarify whether monotherapy with a gastromucoprotective drug, geranylgeranylacetone (GGA), inhibits NSAID-induced acute mucosal injury of the upper digestive tract and small intestine. Methods A prospective, randomized, comparative study. All procedures were performed at Osaka Medical College. The subjects, thirty healthy adult volunteers, were randomly divided into two groups. In the NSAID- GGA group, 75 mg/day of diclofenac sodium and 15 mg/day of GGA were orally administered for two weeks. In the NSAID-FAM group, 75 mg/day of diclofenac sodium and 2 mg/day of famotidine (FAM) were orally administered for two weeks. esophagogastroduodenoscopy (EGD) and video capsule endoscopy (VCE) were performed before and two weeks after drug administration. In addition, we measured fecal occult blood reactions and the fecal calprotectin levels. Results No significant differences were observed between the groups in the mean increase in esophageal/ gastroduodenal lesions. The mean increases in the scores in the NSAID-FAM group (NSAID-GGA group) of small bowel lesions were as follows: erythema: 1.93±.67 (.3±.6), erosions: 1.13±.54 (.38±.35), ulcers:.73±.33 (.7±.7) and edema:.53±.44(.7±.7). The scores for erythema and ulcers were significantly lower in the NSAID-GGA group than in the NSAID-FAM group (p=.32 and.165, respectively). Conclusion We compared the prophylactic effects of a mucoprotective drug, GGA, and an H2RA, famotidine, on mucosal injury involving the esophagus to the small intestine related to the two-week oral administration of diclofenac sodium in healthy volunteers. In the upper digestive tract, the prophylactic effects were similar between the two drugs. However, in the small intestine, GGA more markedly inhibited the development of lesions compared to famotidine. Key words: NSAID, geranylgeranylacetone, video capsule endoscopy, esophagogastroduodenoscopy, upper digestive tract mucosal injury, small intestinal injury () () The Second Department of Internal Medicine, Osaka Medical College, Japan Received for publication August 14, 213; Accepted for publication August 3, 213 Correspondence to Dr. Eiji Umegaki, in238@poh.osaka-med.ac.jp 283

2 Introduction Nonsteroidal anti-inflammatory drugs (NSAIDs) induce various lesions in the gastric mucosa, such as erosions, hemorrhagic lesions and gastric ulcers. Recent advances in digestive tract endoscopy, including the development of wireless video capsule endoscopy (VCE) and double-balloon enteroscopy, have facilitated detailed examinations of the entire digestive tract. As a result, it has been shown that NSAIDs induce not only gastric, but also small/large intestinal mucosal injuries, such as erosion and ulcers (1-3). A previous study reported that combination therapy with NSAIDs and proton pump inhibitors (PPIs) for two weeks causes small intestinal mucosal injury in 68% of healthy adults (3); therefore, the mucosal injury induced by NSAIDs is not prevented by PPIs. According to a survey regarding the incidence of NSAID-induced small intestinal ulcers using a database on patients undergoing double-balloon enteroscopy in Japan, ulcerative lesions are observed in 51% of patients orally treated with NSAIDs (4). As gastric acid is closely involved in the pathogenesis of NSAID-induced gastroduodenal mucosal injury, acid secretion-suppressing therapy with PPIs or H2-blockers (H2RAs) is useful for inhibiting the development of such injuries (5). However, in the small intestine, in which there is no influence of acid, various factors, such as the direct actions of NSAIDs (6), the influence of challenging factors, including enteric bacteria and bile acid (7), and a reduction in protective factors related to a decrease in the prostaglandin (PG) level (8, 9), may be involved depending on the certainty of the influence of these factors. However, no treatment to inhibit NSAIDinduced small intestinal lesions has been established. With respect to the inhibition of small intestinal mucosal injury, studies involving animal experiments have been published. However, few studies have examined this issue in humans. One study reported that combination therapy with diclofenac sodium and a PG analog, misoprostol, in the presence of a PPI for two weeks significantly inhibited small intestinal mucosal injury in healthy adults (1). However, in clinical practice, misoprostol is associated with a high incidence of side effects, such as diarrhea, raising issues regarding continuous administration. Geranylgeranylacetone (GGA) is a gastromucoprotective drug with a low incidence of side effects that is employed to treat gastritis/gastric ulcers in areas of Asia, including Japan. Many animal experiments have shown that GGA induces heat shock proteins (HSPs) responsible for cell protection, inhibiting NSAIDrelated acute gastric/small intestinal mucosal injury. However, no studies have investigated whether GGA exhibits effects on the human upper digestive tract or small intestine. In this prospective, randomized study, we examined whether monotherapy with GGA inhibits NSAID-induced acute mucosal injury of the upper digestive tract (esophagus, stomach) and small intestine using an H2RA, famotidine, instead of a PPI as a control agent. Famotidine was demonstrated to be useful for inhibiting the development of esophagitis and gastric/duodenal ulcers in patients receiving low-dose aspirin in a placebo-controlled study (FAMOUS study) (11). Study subjects Materials and Methods The subjects were 3 healthy adult volunteers. Using sealed assignment cards, the subjects were randomly divided into two groups. Concerning the inclusion criteria, men 2 years of age or older without a history of digestive tract disorders from whom written informed consent was obtained were enrolled. We excluded patients with active digestive tract disorders, those in whom stenosis of the digestive tract was suspected, those with a history of abdominal surgery, those receiving H2RAs, PPIs, PG analogues or other agents that influence gastric acid secretion/gastric movement within two weeks before the start of the study, those in whom a cardiac pacemaker or other medical electronic device was implanted and those who were not considered to be eligible by the chief investigator or attending physicians. This study was conducted in accordance with the Helsinki Declaration (1995), and the protocol was approved by the Ethics Review Board of Osaka Medical College (No.658; May 11, 29). Study protocol This was a prospective, randomized, comparative study. Thirty healthy adult volunteers (2 years or older, men) were randomly divided into two groups. In the NSAID- GGA group, 75 mg/day of diclofenac sodium and 15 mg/ day of GGA were orally administered for two weeks. In the NSAID-FAM group, 75 mg/day of diclofenac sodium and 2 mg/day of famotidine were orally administered for two weeks (Fig. 1). The dose of each drug was determined based on the dose approved by the Japanese Ministry of Health and Welfare. The sample size was calculated based on the results of a review of the incidence of NSAIDinduced small intestinal mucosal injury with a VCE. According to several studies, the incidence of NSAID-induced small intestinal mucosal injury ranges from 5% to 7% (1, 12, 13). Furthermore, Niwa et al. investigated rebamipide, a gastromucoprotective agent for GGA, and reported that the incidence of NSAID-induced small intestinal mucosal injury was 8% and 2% in the placebo and rebamipide groups, respectively (14). Assuming that famotidine, which was employed as a control agent in that study, does not affect the small intestinal mucosa, the incidence of NSAID-induced small intestinal mucosal injury was estimated to be 7% and 2% in the famotidine and GGA groups, respectively. Calculating the number of patients required for the two groups using the chi-square test with a significance level (paired) of 5% and a detection power of 8% identified a sample size of 14.3 for both groups. Therefore, we enrolled 15 patients in each group. Esophagogastroduodenoscopy (EGD) and VCE were per- 284

3 Figure 1. Study protocol. formed before and two weeks after the start of drug administration to compare esophageal/gastric/duodenal and small intestinal mucosal injury. In addition, we measured chemical/immunological fecal occult blood reactions and the fecal calprotectin levels, as markers of digestive tract mucosal permeability, before and after drug administration (15). EGD and evaluation of the EGD findings EGD was performed by four digestive endoscopists who were not informed of the patient assignment before or two weeks after the drug administration. Patients with Grade A or higher mucosal breaks of the esophagus were regarded as having abnormal findings according to the Los Angeles Classification. The presence of gastric/duodenal mucosal injuries, such as erosion and ulcers, was evaluated using the Modified Lanza Score (MLS) (16, 17). VCE and evaluation of the VCE findings For VCE of the small intestine, a PillCam TMSB device (Given Imaging, Ltd., Yoqneam, Israel) was employed. To analyze the VCE images, we used the Rapid Reader 4 software program (Given Imaging, Ltd.). VCE is widely used to diagnose small intestinal diseases. In some cases, however, the ability to observe the target sites during VCE is very poor due to residue, etc. For pretreatment prior to VCE, the subjects were fasted from 9: p.m. the day before the examination (only water was ingested), as described by Nouda et al. (18). The subjects received 1 L of polyethylene glycol solution (PEG)(Niflec ; Ajinomoto Pharma Co., Ltd., Tokyo, Japan) containing 2 mg of dimethylpolysiloxane (Baros ; Horii Pharmaceutical Ind., Ltd., Osaka, Japan) over one hour starting at 6: a.m. on the day of the examination, and VCE was then performed at 9: a.m. The small intestine was examined eight hours after capsule administration. Two digestive endoscopists (K.T. and E.U.) who were not informed of the subjects background and are familiar with VCE independently evaluated the VCE images of the small intestine. Small intestinal mucosal injury was assessed based on the number of findings with respect to six types of injury: erythema, hemorrhage, erosion, ulcer, edema and stenosis. Erythema was defined as a red region with a border extending from the peripheral normal mucosa, erosion was defined as a defect of the normal lustrous mucosa and ulcers were defined as mucosal defects covered with white moss based on the classification reported by Fujimori et al. (19) and Niwa et al. (14) with slight modifications. Inflammatory changes in the small intestinal mucosa were also evaluated using the Lewis score (2) in addition to these six findings. Evaluation of the calprotectin levels To evaluate the grade of NSAID-induced digestive tract mucosal injury, we measured the level of calprotectin, a marker of permeability of the digestive tract due to inflammation, before and after drug administration. Stool samples were collected and stored in a freezer (-2 or lower) within a few hours. The stool samples (5 g) were tested using a Calprotectin ELISA KIT produced by Immunodiagnostik AG (Germany). In brief, 5 g of feces was mixed with 1 ml of extraction solution (Tris buffered isotonic saline with 1 mm of CaCl, ph 8.4) in a high-speed homogenizer. The homogenate was then centrifuged for 2 minutes (1, g) before the supernatant was harvested, and the calprotectin level was measured using the enzyme-linked immunosorbent assay (ELISA). Statistical analysis The age and body mass index (BMI) were compared between the NSAID-GGA and NSAID-FAM groups using Wilcoxon s signed-rank test. To compare the rates of smoking, alcohol consumption, H. pylori infection and fecal occult blood, the chi-square test was employed. Furthermore, the MLS, small intestinal mucosal injury count, Lewis score and calprotectin level before and after drug administration were analyzed using Fisher s PLSD test, with the pretreatment values as baseline values. The data are expressed as the mean±standard deviation (SD). A value of p<.5 was regarded as significant. All statistical analyses were con- 285

4 Table 1. Baseline Characteristics of Subjects No.subjects Age,y, mean ± SD BMI,mean ± SD Current smoking(%) Current alcohol use(%) H.Pylori infection(%) Fecal occult blood test (chemical): positive reaction(%) Fecal occult blood test (immunological):positive reaction(%) NSAID-GGA group NSAID-FAM group p value ± ± 3.2 NS 21.7 ± ± 1.9 NS 2(13.3%) 2(13.3%) NS 2(13.3%) 2(13.3%) 1(6.6%) (%) NS NS 1(6.7%) 1(6.7%) NS 1(6.7%) 1(6.7%) NS Gastric transit time Baseline,mean ± SD 56.3 ± ± 63.7 NS Post-treatment,mean ± SD (min) Small bowel transit time 34.1 ± ± 85.1 NS Baseline,mean ± SD (min) 265. ± ± 8.9 NS Post-treatment,mean ± SD (min) ± ± 85.6 NS BMI:body mass index, MLS:modyfied Lanza score, NS:not significant Table 2. Number of Esophageal/gastroduodenal Lesion at Baseline and Posttreatment Upper Gastrointestinal Endoscopy NSAID-FAM group No.Gastoduodenal lesion (MLS), mean ± SD NSAID-GGA group No.Gastoduodenal lesion (MLS), mean ± SD Baseline Post-treatment p value 1. ± ± ±.52.2 ± *There were no esophageal lesion in either group before and after drug administration. ducted employing the SPSS 16. software package (SPSS Japan, Tokyo, Japan). Baseline characteristics Results There were no significant differences in background factors, such as age, BMI or the rates of smoking, alcohol consumption or H. pylori infection between the NSAID-GGA and NSAID-FAM groups. Furthermore, there were no significant differences in the proportion of patients with a positive reaction to fecal occult blood after drug administration or the duration required for the VCE to pass the stomach/small intestine between the two groups (Table 1). Evaluation of esophageal/gastroduodenal lesions at baseline and after treatment The changes in the MLS on EGD before and after drug administration in the NSAID-GGA and NSAID-FAM groups are shown in Table 2. There were no significant differences between the two groups before drug administration. After drug administration, there was no new development of gastric/duodenal lesions. In this study, there were no esophageal lesions in either group before or after drug administration. Evaluation of small intestinal lesions at baseline and after treatment The VCE findings of typical small intestinal mucosal injuries related to the two-week administration of diclofenac sodium are presented in Fig. 2. The changes in small intestinal mucosal lesions on VCE before and after NSAID administration in the NSAID-GGA and NSAID-FAM groups are shown in Table 3. Before drug administration, there were no significant differences in the number of small intestinal lesions or the Lewis score. When examining the total number of findings regarding the six types of small intestinal mucosal injury, the number of lesions increased with treatment in the NSAID-FAM group, but not in the NSAID- GGA group (NSAID-FAM group: 1.± ±5.65, p=.12, NSAID-GGA group: 1.± ±3.1, p=.344). With respect to the types of small intestinal mucosal in- 286

5 Figure 2. VCE findings of typical small intestinal mucosal injury after treatment. a: erythema: a red region with a border extending from the peripheral normal mucosa. b: erosion: a defect of the normal lustrous mucosa. c: edema. d: ulcers: mucosal defects covered with white moss. The lesions are indicated by a blue circular line. Table 3. Number of Small Intestinal Lesion by Video Capsule Endoscopy at Baseline and Post-treatment NSAID-FAM group No.Small bowel total lesion mean ± SD erythema, mean± SD erosion, mean± SD ulcer, mean± SD edema, mean± SD Lewis score, mean± SD NSAID-GGA group No.Small bowel total lesion mean ± SD erythema, mean± SD erosion, mean± SD ulcer, mean± SD edema, mean± SD Lewis score, mean± SD Baseline Post-treatment p value 1. ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

6 Table 4. Calprotectin at Baseline and Post-treatment Baseline Post-treatment NSAID-FAM group Calprotectin, mean ± SD 4,364 ± 5,259 13,199 ± 16,478 NSAID-GGA group Calprotectin, mean ± SD 7,332 ± 11,681 1,265 ± 13,626 p value jury, erythema (.8± ±2.43, p=.34), erosion ( 1.13±2.1, p=.46) and ulcers (.8±1.32, p=.26) were noted after administration in the NSAID-FAM group. However, no lesion development was observed in the NSAID-GGA group. In this study, neither hemorrhage nor stenosis were noted after the administration of NSAIDs in either group. We also evaluated inflammatory changes in the small intestinal mucosa using the Lewis score. In the NSAID-FAM group, a marked increase was observed in the scores after administration. However, there were no increases in the NSAID-GGA group (NSAID-FAM group: 8.± ±18.91, p=.15, NSAID-GGA group: 15.±58.1, p=.326). Evaluation of the calprotectin levels at baseline and after treatment The changes in the calprotectin levels before and after NSAID administration in the NSAID-FAM and NSAID- GGA groups are shown in Table 4. Before drug administration, there were no significant differences between the two groups. After drug administration, an increase was observed in the calprotectin levels in both groups; however, the increase was not statistically significant in the NSAID-GGA group (NSAID-FAM group: 4,364±5,259 13,199±16,478, p=.58, NSAID-GGA group: 7,332±11,681 1,265± 13,626, p=.532). Discussion In the present study, a gastromucoprotective drug, GGA, inhibited diclofenac sodium-related esophageal/gastric/duodenal mucosal injury. Its effects were similar to those of an H2RA, famotidine. On the other hand, with respect to diclofenac sodium-related small intestinal mucosal injury, erythema, erosion and ulcer development was observed in the famotidine group. However, GGA inhibited the development of these lesions relative to famotidine, which did not decrease the absolute number of lesions in the small bowel compared to the baseline values. When evaluating the results using the Lewis score, which is important for comparing individual findings of small intestinal mucosal injury on VCE and evaluating inflammatory changes in the small intestinal mucosa, GGA was also found to significantly inhibit the elevation of the score compared to famotidine. Furthermore, as another characteristic of this study, we examined the fecal calprotectin level, which is considered to be a marker of lower digestive tract mucosal injury-related enhancement of permeability, and evaluated the grade of mucosal injury by reviewing capsule endoscopy findings and simultaneously measuring the level of this objective biochemical marker. The fecal calprotectin level is correlated with the Crohn s disease activity index (CDAI) and endoscopic findings in patients with Crohn s disease (21). This parameter is reportedly useful for screening inflammatory enteric disorders and malignant large intestinal disorders (22). In this study, GGA significantly inhibited an increase in the fecal calprotectin levels in comparison with famotidine in the patients with diclofenac sodium-related digestive tract mucosal injury. The VCE findings, Lewis score and calprotectin level each changed in parallel. These markers are not specific for NSAID-associated small intestinal lesions, but rather reflect inflammatory changes in the small intestine induced by NSAIDs. The finding that GGA inhibited NSAID-induced small intestinal mucosal injury in this study, different from the effects of famotidine, is highly reliable. Therefore, an H2RA that inhibits acid secretion, famotidine, did not prevent NSAID-induced small intestinal mucosal injury, although some gastromucoprotective agents prevent such injury. In humans, results showing that a PPI, omeprazole, does not inhibit NSAID-induced small intestinal injury have been reported. However, the present report demonstrated that famotidine, as an H2RA, does not act on the small intestine. A placebo-controlled study (FAMOUS study) showed that famotidine is useful for inhibiting the development of esophagitis and gastric/duodenal ulcers in patients receiving low-dose aspirin. When administering NSAIDs, including aspirin, physicians must pay attention to not only the upper digestive tract, but also the lower digestive tract, including the small intestine. It is recommended that when administering NSAIDs, acid secretion inhibitors should be simultaneously administered as first-choice agents. A recent report found that acid suppressants, including PPIs, may exacerbate NSAID-induced small intestinal injury by inducing dysbiosis in animals (23). We cannot recommend the use of acid suppressants, considering the risk of small intestinal mucosal injury. Previous studies have also reported that anti-ulcer agents, such as misoprostol (1), rebamipide (14) and GGA (24), inhibit the development of small intestinal mucosal injury related to short-term NSAID therapy in healthy adults. In these studies, combination therapy with PPIs was administered in both the test agent and placebo groups, focusing only on the small intestine. The present study investigated 288

7 the prevention of mucosal lesions involving the esophagus, stomach and small intestine at the time of NSAID administration. The results suggest that monotherapy with gastromucoprotective agents, such as GGA, prevents the development of lesions of the esophagus to the small intestine and that combination therapy with PPIs may not be necessary for patients without a past history of esophagitis or peptic ulcers. In research, GGA is commonly employed as an HSPinducing agent. HSP promotes protein folding/membrane passage and protects cells against various types of stressors, such as infection and inflammation. Organs/cells in which HSP has been induced show a potent activity to resist toxic substances (25, 26). In mice, HSP induction inhibits NSAID-related gastric mucosal injury (27). In humans, GGA inhibits NSAID-related gastric mucosal injury by inducing HSP (28). In the intestine, GGA inhibits the development of lesions, such as ischemia/reperfusion-related small intestinal mucosal injury, in rats by inducing HSP (29). Recently, a study using laboratory animals suggested that the HSP induction-associated inhibition of inflammatory responses and apoptosis is involved in the mechanism of action of GGA in NSAID-induced small intestinal mucosal injury (3). Misoprostol, an agent in the same class as GGA, has been demonstrated to be useful for inhibiting NSAID-induced digestive tract lesions. However, the incidence of side effects, such as diarrhea, is higher than that for other agents, and the continuation of treatment is often affected. However, GGA did not cause any symptoms or abnormalities in laboratory data during the study period. This study demonstrated that the results obtained in humans are similar to those observed in laboratory animals. In this study, healthy volunteers were enrolled and the duration of medication was relatively short. We cannot extrapolate the results of this study to patients requiring PPI administration, for example, those with a history of peptic ulcers, digestive tract hemorrhage at the time of NSAID administration or reflux esophagitis, in whom the risk of upper digestive tract disorders is high. The issue that requires a new study is the use of GGA in patients with a history of upper digestive tract disorders. In conclusion, we compared the prophylactic effects of a mucoprotective drug, GGA, and an H2RA, famotidine, on mucosal injury involving the esophagus to the small intestine related to the two-week oral administration of diclofenac sodium in healthy volunteers. In the upper digestive tract, the prophylactic effects were similar between the two agents. However, in the small intestine, GGA more markedly inhibited the development of lesions compared to famotidine. Author s disclosure of potential Conflicts of Interest (COI). Eiji Umegaki: Honoraria, AstraZeneca Co, Daiichi Sankyo Co, Takeda Co and TOP Co. Kazuhide Higuchi: Honoraria, Eisai Co, AstraZeneca Co and Daiichi Sankyo Co; Research funding, Eisai Co and Astellas Pharma Inc. References 1. Graham DY, Opekun AR, Willingham FF, Qureshi WA. Visible small-intestinal mucosal injury in chronic NSAID users. Clin Gastroenterol Hepatol 3: 55-59, Sidhu R, Sanders DS, McAlindon ME, Kapur K. Capsule endoscopy for the evaluation of nonsteroidal anti-inflammatory druginduced enteropathy: United Kingdom pilot data. Gastrointest Endosc 64: 135, Maiden L, Thjodleifsson B, Seigal A, et al. Long-term effects of nonsteroidal anti-inflammatory drugs and cyclooxygenase-2 selective agents on the small bowel: a cross-sectional capsule enteroscopy study. Clin Gastroenterol Hepatol 5: , Matsumoto T, Kudo T, Esaki M, et al. Prevalence of non-steroidal anti-inflammatory drug-induced enteropathy determined by double-balloon endoscopy: a Japanese multicenter study. Scand J Gastroenterol 43: , Yeomans ND, Tulassay Z, Juhász L, et al. A comparison of omeprazole with ranitidine for ulcers associated with nonsteroidal antiinflammatory drugs. Acid Suppression Trial: Ranitidine versus Omeprazole for NSAID-associated Ulcer Treatment (ASTRO- NAUT) Study Group. N Engl J Med 338: , Somasundaram S, Rafi S, Hayllar J, et al. Mitochondrial damage: a possible mechanism of the topical phase of NSAID induced injury to the rat intestine. Gut 41: , Evans SM, Whittle BJ. Interactive roles of superoxide and inducible nitric oxide synthase in rat intestinal injury provoked by nonsteroidal anti-inflammatory drugs. Eur J Pharmacol 429: , Sigthorsson G, Simpson RJ, Walley M, et al. COX-1 and 2, intestinal integrity, and pathogenesis of nonsteroidal anti-inflammatory drug enteropathy in mice. Gastroenterology 122: , Higuchi K, Umegaki E, Watanabe T, et al. Present status and strategy of NSAIDs-induced small bowel injury. J Gastroennterol 44: , Fujimori S, Seo T, Gudis K, et al. Prevention of nonsteroidal antiinflammatory drug-induced small-intestinal injury by prostaglandin: a pilot randomized controlled trial evaluated by capsule endoscopy. Gastrointest Endosc 69: , Taha AS, McCloskey C, Prasad R, Bezlyak V. Famotidine for the prevention of peptic ulcers and oesophagitis in patients taking low-dose aspirin (FAMOUS): a phase III, randomised, doubleblind, placebo-controlled trial. Lancet 374: , Goldstein JL, Eisen GM, Lewis B, Gralnek IM, Zlotnick S, Fort JG. Video capsule endoscopy to prospectively assess small bowel injury with celecoxib, naproxen plus omeprazole, and placebo. Clin Gastroenterol Hepatol 3: , Maiden L, Thjodleifsson B, Seigal A, et al. Long-term effects of nonsteroidal anti-inflammatory drugs and cyclooxygenase-2 selective agents on the small bowel: a cross-sectional capsule enteroscopy study. Clin Gastroenterol Hepatol 5: , Niwa Y, Nakamura M, Ohmiya N, et al. Efficacy of rebamipide for diclofenac-induced small-intestinal mucosal injuries in healthy subjects: a prospective, randomized, double-blinded, placebocontrolled, cross-over study. J Gastroenterol 43: , Røseth AG, Fagerhol MK, Aadland E, Schjønsby H. Assessment of the neutrophil dominating protein calprotectin in feces. A methodologic study. Scand J Gastroenterol 27: , Lanza FL, Royer GL, Nelson RS, Chen TT, Sechman CE, Rack MF. A comparative endoscopic evaluation of the damaging effectsof nonsteroidal anti-inflammatory agents on ther gastric and duodenal mucosa. Am.J.Gastroenterol 75: 17-21, Naito Y, Yoshikawa T, Iinuma S, et al. Rebamipiode protects against indomethacin-induced gastric mucosal injury in healthy volunteers in double-blind, placebo-controlled study. Dig Dis Sci 289

8 43 (9 Suppl): 83S-89S, Nouda S, Morita E, Murano M, et al. Usefulness of polyethylene glycol solution with dimethylpolysiloxanes for bowel preparation before capsule endoscopy. J Gastroenterol Hepatol 25: 7-74, Fujimori S, Gudis K, Takahashi Y, et al. Distribution of small intestinal mucosal injuries as a result of NSAID administration. Eur J Clin Invest 4: 54-51, Gralnek IM, Defranchis R, Seidman E, Leighton JA, Legnani P, Lewis BS. Development of a capsule endoscopy scoring index for small bowel mucosal inflammatory change. Aliment Pharmacol Ther 27: , Sipponen T, Savilahti E, Kolho KL, Nuutinen H, Turunen U, Färkkilä M. Crohn s disease activity assessed by fecal calprotectin and lactoferrin: correlation with Crohn s disease activity index and endoscopic findings. Inflamm Bowel Dis 14: 4-46, von Roon AC, Karamountzos L, Purkayastha S, et al. Diagnostic precision of fecal calprotectin for inflammatory bowel disease and colorectal malignancy. Am J Gastroenterol 12: , Wallace JL, Syer S, Denou E, et al. Proton pump inhibitors exacerbate NSAID-induced small intestinal injury by inducing dysbiosis. Gastroenterology 141: , Niwa Y, Nakamura M, Miyahara R, et al. Geranylgeranylacetone protects against diclofenac-induced gastric and small intestinal mucosal injuries in healthy subjects: a prospective randomized placebo-controlled double-blind cross-over study. Digestion 8: , Hirakawa T, Rokutan K, Nikawa T, Kishi K. Geranylgeranylacetone induces heat shock proteins in cultured guinea pig gastric mucosal cells and rat gastric mucosa. Gastroenterology 111: , Tanaka K, Tsutsumi S, Arai Y, et al. Genetic evidence for a protective role of heat shock factor 1 against irritant-induced gastric lesions. Mol Pharmacol 71: , Suematsu S, Tanaka K, Namba T, et al. A role for HSP7 in protecting against indomethacin-induced gastric lesions. J Biol Chem 284: , Yanaka A, Zhang S, Sato D, et al. Geranylgeranylacetone protects the human gastric mucosa from diclofenac-induced injury via induction of heat shock protein 7. Digestion 75: , Iwai T, Ichikawa T, Kida M, et al. Protective effect of geranylgeranylacetone against loxoprofen sodium-induced small intestinal lesions in rats. Eur J Pharmacol 652: , Asano T, Tanaka K, Yamakawa N, et al. HSP7 confers protection against indomethacin-induced lesions of the small intestine. J Pharmacol Exp Ther 33: , The Japanese Society of Internal Medicine 29