Effects of nonsteroidal, antiinflammatory drugs on gastrointestinal injury and prostanoid generation in healthy volunteers

Gastric acid is the key modulator in the pathogenesis of non-steroidal anti-inflammatory drug-induced ulceration in rats

1. In the present study, we investigated the role of gastric acid (GA) secretion on non-steroidal anti-inflammatory drug (NSAID)-induced ulcerogenesis in vivo. Rats were administered single oral doses of selective cyclo-oxygenase (COX)-1 (SC-560; 2.5 mg/kg), COX-2 (DFU; 25 mg/kg) or non-selective COX (indomethacin; 25 mg/kg) inhibitors. Three groups (basal, histamine-stimulated and histamine with lansoprazole) were pylorus ligated 2 h after inhibitor administration and killed 2 h later. Another group without pylorus ligation received only inhibitors and was killed after 18 h. 2. At 4 h, indomethacin increased the ulcer index (UI) and myeloperoxidase (MPO) activity in basal and histamine-stimulated states, whereas SC-560 only increased MPO activity. Histamine-stimulated, but not basal, GA was further enhanced by indomethacin and SC-560 via increased proton pump expression. Lansoprazole (10 mg/kg) reduced the UI, MPO activity and GA to basal levels with SC-560 and DFU and to near basal with indomethacin. Indomethacin and SC-560 significantly inhibited prostaglandin (PG) E(2), without significantly affecting COX-1 and COX-2 expression. Although DFU inhibited PGE(2) by one-third, it did not affect COX expression. 3. At 18 h, indomethacin significantly increased the UI and MPO activity, whereas PGE(2) synthesis was less inhibited, indicating a return to control levels. In contrast, PGE(2) synthesis was higher than control with SC-560. Furthermore, COX-2 expression was significantly elevated with indomethacin and SC-560, explaining the source of augmented PGE(2) synthesis. Proton pump expression remained elevated, comparable with 4 h levels, with indomethacin and SC-560. However, DFU had no significant effect on the aforementioned parameters. 4. The data suggest that NSAID-induced ulcerogenesis is dependent on the amount of GA secretion derived from increased proton pump expression and requires inhibition of both COX-1 and COX-2.

Uncoupling of intestinal mitochondrial oxidative phosphorylation and inhibition of cyclooxygenase are required for the development of NSAID-enteropathy in the rat

BACKGROUND

The pathogenesis of NSAID-induced gastrointestinal damage is believed to involve a nonprostaglandin dependent effect as well as prostaglandin dependent effects. One suggestion is that the nonprostaglandin mechanism involves uncoupling of mitochondrial oxidative phosphorylation.

AIMS

To assess the role of uncoupling of mitochondrial oxidative phosphorylation in the pathogenesis of small intestinal damage in the rat.

METHODS

We compared key pathophysiologic events in the small bowel following (i) dinitrophenol, an uncoupling agent (ii) parenteral aspirin, to inhibit cyclooxygenase without causing a 'topical' effect and (iii) the two together, using (iv) indomethacin as a positive control.

RESULTS

Dinitrophenol altered intestinal mitochondrial morphology, increased intestinal permeability and caused inflammation without affecting gastric permeability or intestinal prostanoid levels. Parenteral aspirin decreased mucosal prostanoids without affecting intestinal mitochondria in vivo, gastric or intestinal permeability. Aspirin caused no inflammation or ulcers. When dinitrophenol and aspirin were given together the changes in intestinal mitochondrial morphology, permeability, inflammation and prostanoid levels and the macro- and microscopic appearances of intestinal ulcers were similar to indomethacin.

CONCLUSIONS

These studies allow dissociation of the contribution and consequences of uncoupling of mitochondrial oxidative phosphorylation and cyclooxygenase inhibition in the pathophysiology of NSAID enteropathy. While uncoupling of enterocyte mitochondrial oxidative phosphorylation leads to increased intestinal permeability and low grade inflammation, concurrent decreases in mucosal prostanoids appear to be important in the development of ulcers.

Experimental enteropathy in athymic and euthymic rats: synergistic role of lipopolysaccharide and indomethacin

The aim of this study was to investigate the immunologic and microbiological bases of indomethacin enteropathy. Athymic nude and euthymic specific pathogen-free (SPF) rats were reared under conventional or SPF conditions. In each group, indomethacin was given intrarectally for 2 days. Indomethacin enteropathy was evaluated using a previously described ulcer index and tissue myeloperoxidase activity. Both euthymic and athymic nude rats developed intestinal ulcers to the same degree under conventional conditions but no or minimal ulcer under SPF conditions. Pretreatment of conventional rats with intragastric kanamycin sulfate, an aminoglycoside antibiotic, attenuated indomethacin enteropathy in a dose-dependent fashion. Interestingly, when lipopolysaccharide was injected intraperitoneally in kanamycin-pretreated rats, it fully restored enteropathy in these rats in a dose-dependent manner. We confirmed that kanamycin decreased the number of gram-negative bacteria and endotoxin concentration of the small intestine in a dose-dependent fashion. These results indicate that indomethacin enteropathy is bacteria dependent and does not require a T cell function. Synergy between indomethacin and bacterial lipopolysaccharide may play a major role in this enteropathy.

Effects of chronic treatment with indomethacin at clinically relevant doses on intestinal tissue 6-keto prostaglandin F1 alpha and leukotriene B4 level in relation to gastroenteropathy

This study investigated the effects of indomethacin at clinically relevant doses and its chronic usage on intestinal pathology, survival time and intestinal tissue 6-keto prostaglandin F1 alpha and leukotriene B4 level in rats during various periods with different doses. Indomethacin was administered ranging from 0.625 to 5 mg/kg. When used in doses of 0.625 and 1.25 mg/kg, indomethacin caused no apparent intestinal lesions or death during a treatment period of 30 days. On the other hand, all rats died in 7 days when 5 mg/kg of indomethacin was given. Mortality rate reached 53.3% in seven days in the group where 3.75 mg/kg indomethacin was given. The minimal dose of indomethacin, which induced intestinal ulcer and death, was 2.5 mg/kg. The main pathological findings were intestinal ulcers, but no macroscopic and microscopic changes were observed in the stomach. Intestinal tissue 6-keto prostaglandin F1 alpha and leukotriene B4 levels were quantified by enzyme immunoassay after homogenisation and extraction of tissue. In dose-dependent studies, only the dose of indomethacin, 3.75 mg/kg, significantly inhibited intestinal tissue 6-keto prostaglandin F1 alpha levels during seven days application period (197.39 +/- 24.26 vs 383.66 +/- 46.68 ng/g tissue, treatment vs control). 2.5 mg/kg of indomethacin caused no intestinal ulceration on 4th day, however, it significantly inhibited intestinal tissue 6-keto prostaglandin F1 alpha levels on 4th day in time-dependent studies (190.3 +/- 26.62 vs 383.66 +/- 46.68 ng/g tissue, treatment vs control). Neither dose-dependent nor time-dependent indomethacin administration changed intestinal tissue leukotriene B4 level. The results of this study indicated that indomethacin produced enteropathy rather than gastropathy when used chronically in clinically relevant doses in rats. Inhibition of prostaglandin synthesis, which was estimated by quantification of intestinal tissue 6-keto prostaglandin F1 alpha level, seemed not to be a prerequisite for its enteropathic effect.

Review article: Pathophysiology of the intestinal mucosa in inflammatory bowel disease and arthritis: similarities and dissimilarities in clinical findings

Although apparently dissimilar in both structure and function, the gut and locomotor system are linked by a number of clinical syndromes in which both are involved. In inflammatory bowel disease, intestinal mucosal inflammation is the primary problem, but may be associated with both axial and peripheral arthropathies. In the seronegative spondyloarthropathies the primary problem is in the locomotor system, but intestinal abnormalities may also be present. In addition, non-steroidal anti-inflammatory drugs used in their treatment, may cause a distinct enteropathy which may be confused with the disease-related intestinal abnormalities. The clinical, histological and genetic features of these conditions are examined in order to assess their relationship and the central role of the intestinal mucosa in their pathogenesis.

Influence of anti-rheumatic drugs on gut permeability and on the gut associated lymphoid tissue

There is great interest in the association between intestinal inflammation and the various arthropathies. However, most studies assessing intestinal function in these diseases are confounded by the fact that non-steroidal anti-inflammatory drugs (NSAIDs) have profound effects on the small intestine. Hence NSAIDs cause quite distinct and severe biochemical damage during drug absorption (uncoupling of mitochondrial oxidative phosphorylation proving to be most important) which results in increased intestinal permeability. All commonly used NSAIDs, apart from aspirin and nabumetone, are associated with increased intestinal permeability in man. Whilst reversible in the short term, it may take months to improve following prolonged NSAID use. Increased intestinal permeability appears to be the central mechanism of converting the biochemical damage to an inflammatory tissue reaction (NSAID enteropathy). The inflammatory enteropathy is not, however, unique to NSAIDs but similar changes are found with other permeability breakers. In intestinal infections and in diseases associated with reduced mucosal defence, suggesting that the small intestinal inflammation represents a common final pathway for a number of intestinal injuries. Spondylarthropathies are associated with a high prevalence of terminal ileitis, but as most patients have been receiving NSAIDs it has been difficult to dissociate the effects of NSAIDs on intestinal function from that of the ileitis itself. Nevertheless, two studies suggest that increased intestinal permeability in spondylarthropathies occur independently of NSAID ingestion. Whilst these findings may have implications for the development of arthritis, the permeability changes in spondylarthropathy do not differ quantitatively or qualitatively from that of NSAIDs or other permeability breakers. NSAID enteropathy can be differentiated from spondylarthropathic enteropathy by differences in location of disease and lack of predilection of certain HLA types. However, as the two may coexist both enteroscopy and ileocolonoscopy may be necessary for this distinction.

Selective inhibition of fatty acid oxidation in colonocytes by ibuprofen: a cause of colitis?

Ibuprofen is associated with initiation or exacerbation of ulcerative colitis. As ibuprofen selectively inhibited fatty acid oxidation in the liver or caused mitochondrial damage in intestinal cells, its effect on substrate oxidation by isolated colonocytes of man and rat was examined. Ibuprofen dose dependently (2.0-7.5 mmol/l) and selectively inhibited 14CO2 production from labelled n-butyrate in colonocytes from the proximal and distal human colon (n = 12, p = < 0.001). Glucose oxidation was either unaltered or increased. Because short chain fatty acid oxidation is the main source of acetyl-CoA for long chain fatty acid synthesis, the inhibition of prostaglandin synthesis by ibuprofen in the colonic mucosa could also occur at this level. Because the concentrations of ibuprofen that can be attained in the human colon are not known, conclusions drawn from current dosages are tentative. The inhibition of fatty acid oxidation by ibuprofen may be biochemically implicated in the initiation and exacerbation of ulcerative colitis, manifestation of which would depend on the ibuprofen concentrations reached in the colon.

Effects of nonsteroidal antiinflammatory drugs on ulcerogenesis and gastric secretion in pylorus-ligated rat

The effects of nonsteroidal antiinflammatory drugs on ulcerogenesis and gastric secretion were evaluated in a pylorus-ligated rat model. Oral administration of salicylate (50 mg/kg), aspirin (50 mg/kg), and indomethacin (3.5 mg/kg) significantly increased ulcerogenesis over the basal value by six- to sevenfold, but ibuprofen's (10 mg/kg) fourfold increase was not significant. Aspirin in conjunction with histamine (0.5 mg/kg subcutaneously) significantly increased ulcerogenesis by 2.7-fold compared to histamine alone. Basal acid secretion was increased significantly by 156% after indomethacin, but not by other nonsteroidal antiinflammatory drugs. In contrast, all nonsteroidal antiinflammatory drugs, except indomethacin, significantly decreased histamine-stimulated acid secretion. Non-steroidal antiinflammatory drugs had no effect on pepsinogen secretion. Ranitidine pretreatment (25 mg/kg intraperitoneally) significantly decreased basal acid and pepsinogen secretion in all treatment groups by > 85% and > 40%, respectively, and ulcerations induced by salicylate, aspirin, and indomethacin were also inhibited by 90%, 60%, and 60%, respectively. The observed inhibition of prostaglandin E2 generation by nonsteroidal antiinflammatory drugs under basal secretory conditions appeared to correlate with the extent of ulcerogenesis. Our data support the concept that acid, in addition to inhibition of prostaglandin E2 synthesis, plays an important role in the pathogenesis of nonsteroidal antiinflammatory drug-induced gastropathy.

Effects of diet on experimentally induced intestinal ulcers in rats: morphology and tissue leukotrienes

The effects of dietary pretreatment on longitudinal ulcers of the intestine induced by indomethacin given intracolonically were investigated in rats. The rats were pretreated with either standard diet or liquid meals. Intracolonic indomethacin (24 mg/kg/day) given for two days produced longitudinal ulcers and small scattered ulcers in the small intestine in the control rats that were receiving standard pelleted formula. Three days pretreatment with one of two types of liquid meals, low residual diet (LRD) or elemental diet (ED), significantly reduced the incidence (3% in ED group and 0% in the LRD group) and the length of the longitudinal ulcers in the small intestine. The caecum was affected in each dietary pretreatment group (67% in controls, 80% in LRD group, and 69% in ED group). Colonic ulcers that were located in a longitudinal fashion were found in 42% of LRD group, while these ulcers were less frequently found in the ED group (13%) and controls (0%). Development of ulcers in the caecum and in the colon of rats in ED and LRD groups was more delayed than that of small intestinal ulcers of control rats. In another experiment, pretreatment by ED significantly increased colonic tissue leukotriene B4 concentration when compared with that of controls. These findings suggest that the site of experimental enteropathy induced by indomethacin given intracolonically can be modified by dietary pretreatment. This animal model can be available for investigating differences in the pathophysiology of enteropathy according to the site of involvement.

Intestinal toxicity of non-steroidal anti-inflammatory drugs

We review the adverse effect of non-steroidal anti-inflammatory drugs (NSAIDs) on the small and large intestine. NSAIDs cause small intestinal inflammation in 65% of patients receiving the drugs long-term. The clinical implications of NSAID-induced enteropathy are that patients bleed and lose protein from the inflammatory site, contributing to iron deficiency and hypoalbuminemia, respectively. Some patients develop intestinal strictures, which may require surgery, and the occasional one may develop discrete ulcers with perforations. There are a number of therapeutic options available to treat the enteropathy and the attendant complications, including antibiotics, sulphasalazine and misoprostol. The colon, by comparison, is only rarely affected by NSAIDs, but colitis is well recognized and NSAIDs may be an important factor in diverticular complications and the relapse of inflammatory bowel disease. There is an association between NSAID intake and appendicitis in the elderly.

Side effects of nonsteroidal anti-inflammatory drugs on the small and large intestine in humans

BACKGROUND

It is not widely appreciated that nonsteroidal anti-inflammatory drugs (NSAIDs) may cause damage distal to the duodenum. We reviewed the adverse effects of NSAIDs on the large and small intestine, the clinical implications and pathogenesis.

METHODS

A systematic search was made through Medline and Embase to identify possible adverse effects of NSAIDs on the large and small intestine.

RESULTS

Ingested NSAIDs may cause a nonspecific colitis (in particular, fenemates), and many patients with collagenous colitis are taking NSAIDs. Large intestinal ulcers, bleeding, and perforation are occasionally due to NSAIDs. NSAIDs may cause relapse of classic inflammatory bowel disease and contribute to serious complications of diverticular disease (fistula and perforation). NSAIDs may occasionally cause small intestinal perforation, ulcers, and strictures requiring surgery. NSAIDs, however, frequently cause small intestinal inflammation, and the associated complications of blood loss and protein loss may lead to difficult management problems. The pathogenesis of NSAID enteropathy is a multistage process involving specific biochemical and subcellular organelle damage followed by a relatively nonspecific tissue reaction. The various possible treatments of NSAID-induced enteropathy (sulphasalazine, misoprostol, metronidazole) have yet to undergo rigorous trials.

CONCLUSIONS

The adverse effects of NSAIDs distal to the duodenum represent a range of pathologies that may be asymptomatic, but some are life threatening.

Placebo-controlled comparison of piroxicam-beta-cyclodextrin, piroxicam, and indomethacin on gastric potential difference and mucosal injury in humans

The acute gastroduodenal mucosa injury and gastric potential difference (GPD) drops provoked by 14-day administration of 20 mg/day of a new piroxicam formulation (piroxicam-beta-cyclodextrin), 20 mg/day standard piroxicam and 100 mg/day indomethacin were evaluated and compared in a randomized, double-blind, placebo-controlled study carried out on 64 volunteers. Endoscopic examinations, performed after 14-day treatment, demonstrated that piroxicam-beta-cyclodextrin was less gastrolesive (mean endoscopic score +/- SE = 0.56 +/- 0.2) than either piroxicam (2.06 +/- 0.5) or indomethacin (2.25 +/- 0.5) (p < 0.01). The drop in GPD after a single dose of the assigned drug was considerably greater for piroxicam and indomethacin than for piroxicam-beta-cyclodextrin (p < 0.01), which registered similar values to placebo. Since GPD is an expression of the anatomo-functional integrity of the gastric barrier, the results indicate that piroxicam-beta-cyclodextrin exerts less direct acute damage on the gastric mucosa. Therefore, when administered short-term, piroxicam-beta-cyclodextrin appears to be less gastrolesive than either indomethacin or the standard piroxicam formulation.

Nonsalicylate nonsteroidal antiinflammatory drugs augment prestimulated acid secretion in rabbit parietal cells. Investigation of the mechanisms of action

The effects of nonsalicylate nonsteroidal antiinflammatory drugs on acid secretion were studied in isolated rabbit parietal cells. Indomethacin, naproxen, and carprofen (10(-6)-10(-4) mol/L) potentiated histamine-, forskolin-, 3-isobutyl-1-methylxanthine-, and dibutyryl cyclic adenosine monophosphate-stimulated acid secretion without affecting basal acid secretion. This augmentation of secretagogue-stimulated acid secretion was dependent on extracellular calcium because potentiation was abolished by calcium depletion of the medium or in the presence of the calcium antagonist lanthanum chloride. Potentiation was independent of the H2 and muscarinic receptors and did not appear to involve guanine nucleotide regulatory proteins. Proton pump activity was unaffected by indomethacin. Furthermore, nonsteroidal antiinflammatory drugs increased calcium efflux through the plasma membrane, as measured by calcium 45, and decreased endogenous prostaglandin E2 content. Exogenous dimethyl prostaglandin E2 inhibited the potentiating effect of these drugs on histamine-stimulated but apparently not on dibutyryl cyclic adenosine monophosphate-stimulated acid secretion. The data indicate that nonsalicylate nonsteroidal antiinflammatory drugs interacted at a postreceptor site between adenylate cyclase and the proton pump. The potentiating effects of these drugs were regulated by calcium and possibly modulated by prostanoids.

Aspirin potentiates prestimulated acid secretion and mobilizes intracellular calcium in rabbit parietal cells

The effects of aspirin on gastric acid secretion were studied in isolated rabbit parietal cells (PC). Aspirin (10(-5) M) potentiated histamine-, dibutyryl cyclic AMP (dbcAMP)-, forskolin- and 3-isobutyl-1-methylxanthine-stimulated acid secretion without affecting basal acid secretion. Augmentation of secretagogue-stimulated acid secretion by aspirin was dependent on calcium (Ca2+) since potentiation was blocked by removal of extracellular Ca2+ ([Ca2+]o) or addition of the calcium antagonist lanthanum chloride. Using the Ca2+ probe fura-2, aspirin (10(-6) - 2 X 10(-5) M) rapidly increased intracellular free Ca2+ concentration ([Ca2+]i) in a dose-dependent manner. The source of released Ca2+ was intracellular as demonstrated by depletion of intracellular Ca2+ and [Ca2+]o with EGTA washing. Aspirin did not affect several other signal transduction sites involved in stimulus-secretion coupling, including the H2 receptor, intracellular cyclic AMP (cAMP), inositol 1,4,5, triphosphate (IP3) and H+,K(+)-ATPase. Aspirin decreased PC prostaglandin E2 (PGE2) content by 98%. Exogenous dimethyl PGE2 (dmPGE2) inhibited both histamine-stimulated acid secretion and its enhancement by aspirin. In contrast, dmPGE2 abolished aspirin-induced potentiation of dbcAMP-stimulated acid secretion by augmenting the dbcAMP-stimulated response. These results indicate that aspirin acts at a site beyond the adenylate cyclase/cAMP system and before the proton pump, presumably by releasing Ca2+ from an IP3-independent intracellular storage pool and by inhibiting PGE2 generation.

Prostaglandins and gastric ulcers: from seminal vesicle to misoprostol (Cytotec)

Misoprostol (Cytotec, G.D. Searle & Company, Chicago, IL) is the first of a new class of orally administered prostaglandin analog drugs to be marketed in the United States. Misoprostol was approved for the prevention of gastric mucosal ulcers associated with nonsteroidal anti-inflammatory drugs (NSAIDS) in high-risk patients. This represents a potentially important development in the pharmacotherapy of peptic ulcer disease. The purposes of this article are to review (1) the biochemistry, physiology, and pharmacology of prostaglandins, especially those synthesized by the stomach; (2) the potential role of prostaglandin deficiency in the pathophysiology of gastric ulcer disease; and (3) the role of prostaglandin analogs in the prevention and therapy of gastric ulcer disease and in other conditions. As the mechanism of action of these new drugs differs from that of the histamine H2-receptor antagonists (H2-blockers), prostaglandin analogs will, whenever possible, be compared with the H2-blockers [cimetidine (Tagamet), ranitidine (Zantac), nizatidine (Axid) and famotidine (Pepcid)], currently the cornerstone of peptic ulcer therapy in this country.

Role of endogenous prostaglandins in preventing gastrointestinal ulceration: induction of ulcers by antibodies to prostaglandins

Active immunization of rabbits with the principal, endogenous prostaglandins in the gastrointestinal mucosa induces gastrointestinal mucosal ulceration. Development of ulceration in prostaglandin-immunized rabbits appears to be a direct consequence of production of specific prostaglandin antibodies, as prostaglandin antibodies per se induce gastric ulceration within 9 days when administered intravenously to unimmunized rabbits. These studies suggest that endogenous prostaglandin E2, F2 alpha, D2, and I2 in the gastrointestinal tract play an important role in preventing mucosal ulceration. The mechanism of ulcer formation is not completely understood, but most evidence points toward prostaglandin antibodies inducing mucosal ulceration by binding to endogenous prostaglandins within the mucosa and thereby negating their mucosal protective effects. Gastric acid hypersecretion and complement fixation by prostaglandin-antiprostaglandin complexes are not likely involved in the development of mucosal ulceration in this model. Use of antibodies to interfere with prostaglandin action may be an alternative approach to investigate (a) the importance of endogenous prostaglandins in mediating mucosal protective mechanisms and (b) the role of prostaglandins in acute and chronic erosive/ulcerative diseases of the gastrointestinal tract.

Mechanisms of gastric damage by non-steroidal anti-inflammatory drugs

All non-steroidal anti-inflammatory drugs (NSAIDs) used in the treatment of rheumatic diseases may cause gastrointestinal mucosal injury. The mechanisms by which these agents damage mucosa are not fully understood, although, reduction of mucosal defence by the depletion of endogenous, protective prostaglandins has been deemed important. NSAIDs have been shown to decrease the magnitude of the mucus-bicarbonate barrier, disrupt the epithelial cell layer, reduce the surface hydrophobicity of epithelial cells and to diminish mucosal blood flow. Such effects render the mucosa more susceptible to damage by acid, pepsin, bile salts and alcohol. In addition, direct mucosal injury may be caused by the physiochemical properties of NSAIDs, being weak acids. There is now increasing evidence that gastroduodenal mucosa adapts to acute damage by these drugs with mucosal injury recovering during continued administration. The mechanisms governing such adaptation remain unknown and require further investigation.