Pathogenesis of nonsteroidal anti-inflammatory drug gastropathy ...

Pathogenesis of nonsteroidalanti-inflammatory druggastropathy: Clues topreventative therapy

Salim MA Bastaki ��, John L Wallace ��

To develop better strategies for preventing the gastric ul-cers and episodes of bleeding that occur in patients who

are chronically ingesting nonsteroidal anti-inflammatorydrugs (NSAIDs), it is necessary to understand the mecha-nisms through which these drugs produce damage to thestomach. There are two major components to the ulcero-genic effects of NSAIDs in the stomach: topical irritant ef-fects on the epithelium and suppression of gastric prostaglan-

din synthesis. Effects of NSAIDs on the hydrophobic natureof the gastric mucosa and on gastric acid secretion may alsocontribute to their ulcerogenic effects.

TOPICAL IRRITATION OF THE MUCOSATopical irritant properties are predominantly associatedwith acidic NSAIDs, such as those with a carboxylic acidresidue. Acetylsalicylic acid is the best characterized NSAID

��

SMA Bastaki, JL Wallace. Pathogenesis of nonsteroidal anti-inflammatory drug gastropathy: Clues to preventative therapy.Can J Gastroenterol 1999;13(2):123-127. ��

��

��

��

��

��

��! ��

��! �� !

�� ! ��

�� " �� "��

��

��

�� "�� !

�� "�� "��

�� #��

�� "��

��

�� "�� $� "��

�� %��

Key Words: ��

La pathogenèse de la gastropathie liée auxAINS : Conseils pratiques pour un traitementpréventif

��

��

�� !" � �� #��

��

�� " ��

�� $ ��

�� % ��

�� % � �� &� ��

�� % � ��

�� #��

��" � ��

�� ' ��

� ��% �� (��" �� %

�� #�� #��

��

��" )��

� �� #�� #�� #��

�� " *� �� #��' ��

�� '��

�� "

�� ! " �� ###$ �� "

� �� %� &�'( �� &$#�%%$�&)#* � + &$#�%,$�##)# �� - �� ./�� (�

NSAID GASTROENTEROPATHY

in terms of topical irritant effects on the gastric epithelium.Epithelial damage caused by NSAIDs is likely related in partto the accumulation of these drugs inside these cells (‘iontrapping’). The un-ionized forms of these drugs can enter theepithelial cells, but, once in the neutral intracellular envi-ronment, they are converted to an ionized state and cannotdiffuse out. Osmotic forces then pull water into the cell, re-sulting in swelling, sometimes to the point of lysis (1,2).

Topical irritant properties have also been attributed tothe ability of NSAIDs to decrease the hydrophobicity of themucous gel layer in the stomach. Lichtenberger (3)and God-dard et al (4) have proposed that this layer is a primary bar-rier to acid-induced damage in the stomach. They have dem-onstrated that a hydrophobic layer is present on the surfaceof the stomach and that it can be altered by various pharma-cological agents. For example, they demonstrated that NSAIDsassociate with the surface active phospholipids within themucous gel layer, thereby reducing its hydrophobic proper-ties (4-6). Indeed, they showed that the mucous gel layer inthe stomachs of rats and mice given NSAIDs was convertedfrom a nonwettable to a wettable state. This was not just anacute effect because they observed that the reduced hydro-phobicity of the mucous gel layer persisted for several weeksto months after cessation of NSAID administration (3,5).

It has been repeatedly suggested that another mechanismthrough which NSAIDS may damage the stomach is via un-coupling of oxidative phosphorylation in epithelial cells(2,7). This theory has been challenged, however, based onthe observation of gastric and duodenal ulcers following par-enteral or rectal administration of NSAIDs (8,9). Erosionsand ulcers can also be produced in experimental circum-stances in which NSAIDs are administered parenterally(10). It is difficult to comprehend how uncoupling of oxida-tive phosphorylation would occur in gastrointestinal epithe-lial cells but not in the numerous other cells that an NSAIDwould have contact with after its absorption.

Attempts have been made to produce NSAIDs with re-duced topical irritant effects. These include formulation ofslow release or enteric-coated tablets, or preparation of thedrug as a prodrug that requires hepatic metabolism in orderto be active (eg, sulindac and nabumetone). However, theincidence of gastroduodenal ulceration with these prodrugsis comparable with that seen with standard NSAIDs (11-13). This is evidence that the topical irritant properties ofNSAIDs are not paramount in their ability to induce frankulceration in the stomach. However, this possibility cannotbe completely excluded. Most NSAIDs are excreted in bile.Therefore, even with parenteral administration of NSAIDs,reflux of duodenal contents into the stomach would result intopical exposure of the gastric mucosa to these drugs.

INHIBITION OF PROSTAGLANDIN SYNTHESISThere is substantial evidence that the ability of an NSAIDto cause gastric damage correlates well with its ability to sup-press gastric prostaglandin synthesis (14-16). There is also agood correlation between the time- and dose-dependency ofsuppression of gastric prostaglandin synthesis by NSAIDs

and their ulcerogenic activity (14,15,17). Why does suppres-sion of gastric prostaglandin synthesis lead to mucosal in-jury? Several components of gastric mucosal defence aremediated by prostaglandins, including the secretion of mu-cus and bicarbonate by the gastric and duodenal epithelium,mucosal blood flow, epithelial cell proliferation, epithelialrestitution and mucosal immunocyte function (18). Inhibi-tion of prostaglandin synthesis, alone, may not result in theformation of gastric erosions or ulcers. However, suppressionof mucosal prostaglandin synthesis leads to a reduction inthe ability of the gastric mucosa to defend itself against lumi-nal irritants. This has been demonstrated in experimentalanimals. Doses of NSAIDs that did not cause gastric injurywere able to greatly increase the susceptibility of the gastricmucosa to damage induced by irritants (eg, bile salts) (19).

EFFECTS ON THE MICROCIRCULATIONThe ability of NSAIDs to reduce gastric mucosal blood flowhas been recognized for several decades (20-22). More re-cently, it has become clear that damage to the vascular endo-thelium is an early event following administration of NSAIDsto experimental animals (23,24). Studies performed in ourlaboratory and others showed that NSAID administration torats resulted in a significant increase in the number of neu-trophils adhering to the vascular endothelium in both themesenteric and gastric microcirculation (16,25-27). Thiswas observed within 30 mins of NSAID administration,which is consistent with the period of time required for sig-nificant inhibition of prostaglandin synthesis by these drugs(28). The contribution of neutrophils to the gastric injury as-sociated with NSAIDs was demonstrated by the observationthat the severity of damage was substantially reduced in neu-tropenic rats (ie, rats pretreated with antineutrophil serumor methotrexate) (24,29). Depletion of neutrophils also re-sulted in prevention of NSAID-induced vascular endothe-lial damage (24). Further evidence for a role of neutrophilsin the pathogenesis of NSAID-induced gastric injury camefrom studies in which monoclonal antibodies against neutro-phil or endothelial adhesion molecules were employed toprevent neutrophil adherence; these antibodies also mark-edly reduced the severity of gastric damage caused byNSAIDs (27,30).

There are two likely pathways through which neutrophiladherence may contribute to the pathogenesis of NSAID-induced gastric mucosal injury (31). First, the adherence ofneutrophils to the vascular endothelium is likely accompa-nied by activation of these cells, leading to the release of pro-teases (eg, elastase and collagenase) and oxygen-derived freeradicals (eg, superoxide anion). These substances may medi-ate much of the endothelial and epithelial injury caused byNSAIDs. Indeed, several compounds that scavenge foroxygen-derived free radicals were shown to reduce signifi-cantly the severity of NSAID-induced gastric mucosal injuryin rats (32,33). The second pathway through which neutro-phil adherence to the vascular endothelium may contributeto mucosal injury is by producing an obstruction of capillar-ies, thereby reducing gastric mucosal blood flow. Interest-

��

��

ingly, the well characterized ability of NSAIDs to reducegastric blood flow (20-22) was shown to occur after the ap-pearance of ‘white thrombi’ in the gastric microcirculation(21).

INTERFERENCE WITH REPAIR OFMUCOSAL INJURY

When the epithelium of the stomach is damaged, it can berepaired within minutes through rapid migration of healthycells from the gastric pits (34). These cells move along thedenuded basement membrane, which is the template neces-sary for the process of ‘restitution’ (34-37). The factors thattrigger the movement of cells over the denuded basementmembrane are not clear. Damage to the basement membranecan occur when it is exposed to acid (28,38,39). This doesnot occur during normal circumstances because of the for-mation of a microenvironment in which the pH is main-tained at near 7 over sites of injury, even in the presence of asignificant acid load in the lumen (28,38). A ‘cap’ of mucus,cellular debris and plasma proteins forms within seconds ofgastric epithelial injury, trapping the plasma that leaks fromthe underlying microcirculation (38). This plasma accountsfor the near neutral pH within the protective ‘mucoid cap’.Even a very brief cessation of mucosal blood flow results in arapid decrease in the pH within the mucoid cap, leading tothe formation of hemorrhagic erosions (28).

One way that NSAIDs can promote gastric injury is by in-terfering with the process of restitution. Following systemicadministration of an NSAID, the pH within the mucoid capover sites of epithelial damage begins to decline in parallelwith the inhibition of prostaglandin synthesis (28). This inturn leads to the formation of hemorrhagic erosions (28).This effect could be prevented through luminal delivery ofexogenous prostaglandins (28). It is possible that this effectof NSAIDs is related to their ability to reduce gastric bloodflow, but this has not yet been established. Interestingly,however, the decrease in pH within the mucoid cap could beprevented by topical application of exogenous prosta-glandins; this treatment would also prevent the formation ofhemorrhagic erosions (28).

ROLE OF ACIDThe observation that NSAID-induced ulcers can occur inachlorhydric individuals (40,41) has contributed to a widelyheld belief that these lesions occur independently of thepresence of acid in the stomach. This was reinforced by sev-eral reports that demonstrated that treatment with hista-mine H+ receptor antagonists did not reduce the incidence ofNSAID-induced ulceration (42-44). This interpretation ofthe latter findings has been quite controversial, however,mainly because of inconsistencies from one study to anotherin the discrimination between gastric erosions and true ul-cers. Many studies have demonstrated that H+ receptor an-tagonists and proton pump inhibitors can prevent NSAID-induced gastric lesions, but prevention of the formation ofthe clinically more significant ulcers, as well as ulcer compli-cations, was not clearly established. Taha et al (45) recently

reported that a high dose of famotidine (40 mg bid) was ef-fective in preventing NSAID-induced ‘ulcers’. It should benoted that the dose required to induce this effect was doublethe dose approved for the healing of ulcers. Moreover, thestudy of Taha et al (45) has been criticized on the basis ofthe definition of ulcer used (break in the mucosa of morethan 3 mm). Graham (46) suggested that if Taha and col-leagues were to reassess their data using a more robust defini-tion for ulcer, they might find that famotidine was ineffec-tive in preventing ulceration, as had been reported in anearlier study (44). On the other hand, the recent demonstra-tion that omeprazole can significantly reduce the incidenceof NSAID-induced ulcers (47) adds support to the claim ofTaha et al (45) that profound suppression of acid secretion, asis produced by omeprazole or a high dose of famotidine, wasnecessary to have a significant impact on the incidence ofNSAID-induced ulcers.

There are several mechanisms through which acid maycontribute to ulcer formation in NSAID users. First, it isclear that acid can exacerbate damage to the gastric mucosainduced by other agents. For example, acid can convert re-gions of ethanol-induced vascular congestion in the mucosato actively bleeding erosions (34). A second mechanismthrough which acid can contribute to ulcer formation is byinterfering with hemostasis. Platelet aggregation, for exam-ple, is inhibited at a pH of less than 4 (48). Third, acid canconvert superficial injury to deeper mucosal necrosis inter-fering with the process of restitution. As described in moredetail above, the basement membrane is crucial to the resti-tution process in that it acts as a template for epithelial cellmigration. If the basement membrane is exposed to acid, it israpidly degraded (39). A fourth mechanism through whichacid contributes to ulcer formation is by inactivation ofgrowth factors. Many growth factors that are important formaintenance of mucosal integrity and for repair of superficialinjury are acid labile (49).

The effects of NSAIDs on gastric acid secretion are perti-nent to the potential role of acid in the pathogenesis ofNSAID-induced gastric ulceration. Prostaglandins exert in-hibitory effects on parietal cells (50). By inhibiting prosta-glandin production by the mucosa, NSAIDs can increasegastric acid secretion (51). Whether NSAID-induced in-creases in gastric acid secretion contribute to the develop-ment of gastric ulcers has not been directly examined.

FUTURE DIRECTIONSA great deal has been learned about the pathogenesis ofNSAID-induced gastric injury over the past two decades. Asa result, a number of new NSAIDs that will have greatly re-duced gastrointestinal toxicity relative to the current drugsare on the verge of being introduced into the marketplace.For example, selective inhibitors of cyclo-oxygenase-2,which will spare the major form of prostaglandin synthasein the gastrointestinal tract, will produce substantially lessinjury than the current NSAIDs that inhibit cyclo-oxygenase-1 and -2 (52). Whether these agents prove to beas effective for the treatment of the full range of inflamma-

��

��

tory conditions currently treated with NSAIDs remains tobe seen, and some experimental studies suggest that this maynot be the case (53). Nitric oxide-releasing NSAIDs provideanother approach to reducing the adverse effects normallyassociated with NSAIDs (54).

Until the newer NSAIDs are introduced and fully ac-cepted by patients with inflammatory diseases, a reduction ofthe incidence of gastric ulcers associated with the use ofthese drugs can be accomplished through profound suppres-sion of acid secretion, either with proton pump inhibitors orwith high doses of histamine H+ receptor antagonists.

ACKNOWLEDGEMENTS: �� &�� '�� (��

)�� )�� *�� +��

�� '�� (��

REFERENCES," -�� " .�/ �� & ��

�� 0 1�� #�� 2�� ,345$,67+8,�4"+" �� % .&�� .% 9�� % :��/�� ;2%

2�� ; % <=�� " >�� & �� 7 ��#��/ �� &�� " �� ; ?� �� ,338$@67+43�33"

@" �� 2" >�� &�� " �� 9�# ��&�� ,338$8578A8�4@"

B" ?�� ;% .�� <�% �� 2" �� & �� &�� &0 �� ; ��&��,345$+8+7?B+,�@6"

8" �� 2% :� C2% 9�� ;;% � �" �� & �� ! �� /� � D/� ��7 �� #�� =��&" � 2�� ,338$,7,8B�4"

A" �� 2% :� C2% ?�� 2�% 9�� ;;% <�� ;1"E�� '�� &�� &"?� �� & ,338$,647�,B3" �� !

5" 2�� >% :��/�� ;2% �� % <=�� " 2� �� '&� �� &"��& ,33A$B+7,43�3B"

4" E� �� % -�� :% .� �� .% <� /��F 1 " ?� �� % �� /� � �� =�� F� ��"�� ,33@$+57B+�@"

3" .��& �% �� % >�� 1" G�� & �� F �� =�� & ��" ?� �� & ,33@$,67,654�44"

,6" :�� ;�% 2�H�� ?:" 1�� D �� & �� "1� ; ��&�� ,33@$5,7BB5�8+"

,," ?�� I% �� ;�% .�� ?�% �#�� 9*" �� & �� '�� " 1�� >�� ,348$@47A8�56"

,+" 1�� ;�% �� <�% �� H�% :�� % 2�� " >�� #�� '�� & �� & ��"�� 2�� ,345$,B57,68B�3"

,@" ./ �� <% 2�� I9% 1�� >% ./F�& 1;" �� 7 ��#�� & �� &� ��" <� ; 1�� ,33,$@+755�4@"

,B" ��D -�" � ��#��/ �� =��& �� #�� #�� & ��" �� ; ?� �� ,343$,A@7+B�@,"

,8" 9�� H�% :�� 1" 9�� & � �� & �� " �� ,34+$+57A+B�@8"

,A" :�� ;�% 2�1�� & ��2% 1� �� % 2�H�� % :% ��

>�� #� �� &� �� & ��'��7 � �� & /� �� &" ?� �� &,33@$,687,A@6�A"

,5" :�� <;9" >�� /�� &��'&�� % � �� & �� &�� &� ��% ��

� �� & �� "?� �� & ,34,$4673B�4"

,4" :�� ;�% >��& �:" ��/ �� " �� >�� ,338$37++5�@8"

,3" :�� <;9" 2�� &�� & �� % �� "<� ; �� ,355$A67B88�A6"

+6" ��& :% �� % 1�� I" -�� / �� " �� ; ��,348$,B378@�3"

+," H� �� >% ?� � �." E�� &�� " ?� �� &,345$3@74,6�5"

++" ?� >;% .��/&�� 9% H�� ;" -�� / �� " �� ,345$+687@33�B6@"

+@" 9�� H�" 2��#�� =��& �� &� �� & �� " �� ,34@$,@7B85�A6"

+B" :�� ;�% H�� 12% ?�� " ?� �� &�� & �� " �� ; ��&�� ,336$+837?BA+�5"

+8" ��F� .% H�� % :�� ;�% ?�� >% :�� 9E% ?�� "�� F��& � �� #��7�� '&�� " �� ; ��&�� ,33+$+A+7?36@�4"

+A" ��F� .% H�� % :�� ;�% :�� 9E% ?�� " 2�� F��& � �� #�� & �� &� �"?� �� & ,33+$,6@7,BA�8+"

+5" :�� ;�% 2�H�� :% 2�&�F 2% � �" 9�� =��&" �� ;��&�� ,33@$+A87?33@�4"

+4" :�� ;�% 2�H�� ?:" >�� #�� " � ��. ��#�� &�� & �� "?� �� & ,336$337+38�@6B"

+3" �� 2% �� H% -�� 2" �� =��& �� " �� ; ��&��,33+$+A@7?3+6�A"

@6" :�� ;�% �� H�E% 2�H�� ?:" � �� &�� 1�,4 ��F��& � �� #�� " ?� �� &,33,$,667454�4@"

@," :�� ;�% ?�� " � �� &7�� 0 >�� ,33+$,@7,+3�@,"

@+" �� % -�� % <�� ?% �� 1" *'&�� /� � �� =��&"�� ,348$,57B4B�4"

@@" G�� 2% 2�� ;<% :�� ;�" 9�� '&��#�� =��&" �� ; ��&��,33,$+A,7?B56�8"

@B" 2�� ?�% :�� ;�" >�� " G��/� �� < 1�� 2�� ,34,$@47+@�@4"

@8" #�� H% � � % >F�� H% �� :" 9�� #� �� /� ��&�� " ?� �� & ,34+$4+7,B63�+A"

@A" ��& E9% � � " 9�� =��&" �� #�� ,34B$8,785@�4@"

@5" 2�� 9% 1�� % 2�� ;�" 9�� #� �� =��&" �� #�� ,343$A67+@5�BB"

@4" :�� ;�% :�� <;9" 9�� " �� #��& �& ��& �� " ?� �� & ,34A$3,7A6@�,,"

@3" <��F <�% 2�� ?�% :�� ;�" E�� " G��/� �� < 1�� 2�� ,348$867,63�,4"

B6" 9�� -% >�� E�" 2�� & �� /� �� &��" ; 1�� ?� �� ,33@$,A7+,3�+,"

B," ;�� 2% ��=F�� <�% G��F� ;�% <�� % �� 1<"��&�� " �� ,33B$@37@A+�8"

B+" ��/� �2" E��& �� #�� & �� "<� ; 9�� ,338$@B� �� ,!78�,6"

B@" H�� 2% 1�� % ��D� �% -�� -% �� 1" ��#�� =��&7 �� &��

��

��

�� /� � �� .+�� " �� ,338$,@� �� ,!7A+�5B"

BB" �� >;% <�� 2�% .��#�� 2E% �� % <�� 9?"� �� & �� #�� /� � �� & �� !7 �� ) �� " �� ; ?� �� ,33B$437,ABB" �� !

B8" >� � % .�� % ./F�& 1;% � �" -�� #�� & �� & ��" � E�� ; 2�� ,33A$@@B7,B@8�3"

BA" ?�� I" -�� #�� & � ��"� E�� ; 2�� ,33A$@@87,@++"

B5" ./F�& 1;% H�� ;�% D�D��F� �% � �" *��D�� /� � �� /� � �� & ��" *��D�� #�� 2�� )�� 2�� *2��)2! ��& ?��"� E�� ; 2�� ,334$@@475+5�@B"

B4" ?�� -:% H�� 22% 1�� E% ��#�� ." E�� 7 �� "?� �� & ,354$5B7@4�B@"

B3" D�� % -��F�� ;% G & �% 2�� 9E% ��F�� ? % H � H"�� & �� / � �� " ?� �� &,33B$,6A7,,6A�,,"

86" �� ." 2�� & ��7 � �� " �� ; ?� �� ,34A$+,7,�A"

8," ��F& 2% ?� � I% I�� >" �� & �� " ��,34@$+,37@6,�@"

8+" J�� :% 9�� % �� " 2� �� ?K. �&� ��7 ��/ �� & ��"�� # 9�� ,33+$+87+B3�A8"

8@" :�� ;�% <F �% 2�H�� :% �� % ��F�& H�%2�� :H" 1&��'&��, �� &�� 7 �� ?� �'�� &"?� �� & ,334$,,87,6,�3"

8B" :�� ;�% E�� % �� % 2�H�� :% �� -%1�� ?" ?� �� & ��7 >��#�� '�� " �� # 9��,335$B+7,BB�3"

��

��

Submit your manuscripts athttp://www.hindawi.com

Stem CellsInternational

Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014


MEDIATORSINFLAMMATION

of


Behavioural Neurology

EndocrinologyInternational Journal of



Disease Markers


BioMed Research International

OncologyJournal of



Oxidative Medicine and Cellular Longevity


PPAR Research

The Scientific World JournalHindawi Publishing Corporation http://www.hindawi.com Volume 2014

Immunology ResearchHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

Journal of

ObesityJournal of



Computational and Mathematical Methods in Medicine

OphthalmologyJournal of


Diabetes ResearchJournal of



Research and TreatmentAIDS


Gastroenterology Research and Practice


Parkinson’s Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttp://www.hindawi.com

Pathogenesis of nonsteroidal anti-inflammatory drug gastropathy ...

Documents

Transcript of Pathogenesis of nonsteroidal anti-inflammatory drug gastropathy ...