**4. Management of specific populations**

Management strategies are designed to reduce the incidence of GI complications and take into account patient's overall risk and specific NSAID-related risk factors. Most guidelines

Chronic NSAIDs Therapy and Upper Gastrointestinal Tract – Mechanism of Injury,

management in the very high risk group of patients (Chan et al., 2007).

*Patients requiring high-doses NSAIDs* 

balance the risk/benefit ratio..

Aalykke,*.*1999; Cullen, 1997).

*Patients on corticosteroids* 

*Helicobacter pylori positive* 

Mucosal Defense, Risk Factors for Complication Development and Clinical Management 159

was demonstrated effective for prevention of ulcer re-bleeding (13-month incidence of 0% vs. 8.9% in patients treated with celecoxib alone) and considered the best treatment

When physicians have to prescribe high NSAID doses, there is a significant increase in the relative risk of development of GI complications (about three-fold).(Henry et al., 1996; Langman et al., 1994). In those cases, pharmacological and clinical evidences demonstrated that coxibs are safer than nsNSAIDs with a similar anti-inflammatory and analgesic effects. However, high-doses of coxibs show an overall increased risk of adverse events (both CV and related to fluid retention) and this should be taken into account in each single case to

Large population-based studies and meta-analysis demonstrated that *Hp* infection induce a two-fold increase in the risk of developing peptic complications in patients receiving NSAIDs (Chan et al., 2002; Vergara et al., 2005). Moreover, also in patients receiving coxibs, *H pylori* remain a risk factor for development of GI ulcers and bleeding (Bombardier et al., 2000). Systematic reviews and meta-analysis (Chan, 1997; Chan 2002; Vergara 2005) confirmed the efficacy of *Hp* eradication in preventing upper GI complications in patients on chronic NSAIDs therapy, even though treating *Helicobacter pylori* does not completely abolish the risk of bleeding in high risk patients (Chan, 2001). Therefore, even though it is often underestimated in general clinical practice, a test-and-treat strategy is mandatory in patients who require long-term NSAIDs therapy (Gabriel, 1991; Wolfe, 1999; Sauerbaum, 2002; Barkin, 1998; Laine, 2002; Chan, 2002; Laine, 1992; Hawkey, 1998; Loeb, 1992;

Corticosteroids have a synergistic effect with NSAID, magnifying their GI toxicity, and an intrinsic gastrolesive potential effect, specially in patients with multiple concurrent diseases; the risk of ulcer development is increased both in patients receiving NSAIDs and in non-NSAID users. A correct strategy, for the management of patients who need both corticosteroids and NSAIDs was suggested by post-hoc analysis of large clinical trials,

When prescribing NSAIDs to patients requiring high-doses of corticosteroids, a management strategy able to guarantee a reduced GI risk seems to be the choice of a coxib coupled by gastroprotection with a PPI; however, specific data in this setting are lacking

Co-prescription of anticoagulants and NSAIDs induce a significant increase in GI events and bleeding (both clinically manifest and occult); consequently, in these patients NSAID prescription should be considered a contraindication. Although there is a lack of specific data, in those patients when necessary, the co-prescription of a selective COX-2 inhibitor plus a PPI have to be considered in order to reduce the high risk of bleeding and the high

showing that prescription of coxibs seems to reduce the risk of GI complications.

(Holvoet et al., 1991; Hochain et al., 1995; Laine et al., 2010; Weil et al., 2000).

*Use of anticoagulant agents or patients affected by coagulopathy* 

morbidity and mortality that goes with it.

describe specific strategies for high-risk categories of patients; designed to significantly reduce side effects (both general and gastrointestinal) and the number needed to treat (NNT) to achieve the endpoints (reduction in GI events). These strategies often involve the prescription of a gastroprotecive agent (PPIs), reduction of nsNSAIDs dose and the change to selective COX-2 inhibitors.

#### *Management of low-risk patients*

Most guidelines do not suggest gastroprotective strategies for low-risk patients. Physicians prescribing NSAIDs to low-risk patients (less than 65 years, no comorbidity, no concomitant antiplatelet, anticoagulants or corticosteroids and no previous history of NSAID-related GI complications) should follow the general suggestions for GI complications reduction (eg. prescribe the lowest effective dose and avoid drugs with high GI toxicity).

However, even in patients without risk factors, two clinical trials demonstrated a significant reduction of GI events (detection of asymptomatic ulceration and bleeding) in patients receiving coxibs compared to those on nsNSAIDs (Bombardier et al., 2000; Silverstein et al., 2000); these evidences suggest that the low-risk category of patients could become a "norisk" one, if well managed.

#### *Patient with history of peptic ulcer disease*

Epidemiological and retrospective studies identified a past episode of peptic ulcer as a risk factor for development of GI events in patients receiving chronic NSAID therapy; moreover, both *Hp* positive and negative patients present an increased relative risk of complications (Rockal et al., 1995). In order to reduce the risk of GI injury, the switch to a coxib was evaluated in patients with a past history of ulcer disease; both rofecoxib and celecoxib demonstrated their efficacy in reduction of GI events (from 8.8/100 patient/year with non selective NSAID to 2 with rofecoxib) (Laine, 2001). Therefore, in patients with a previous history of peptic ulcer disease, the switch to a selective COX-2 inhibitor could be considered a practical and cost-effective strategy.

Although switching to a coxib induces a significant risk reduction, in these patients there is still a high residual risk of development of GI complications (10 events per 100 patients treated/year in VIGOR study) (Bombardier et al., 2000). In this setting, the combination therapy of a PPI to a standard NSAID appears to be more appropriate than a coxib alone; clinical studies demonstrated that combination with omeprazole induces a significant reduction in ulcer development both in patients with previous ulceration and perforation when compared to patients receiving a COX-2 selective inhibitors (rofecoxib) or monotherapy with ns-NSAID (ibuprofen) [Cullen et al., 1998; Hawkey et al., 1998). After this first evidence, PPIs other than omeprazole (both lansoprazole, pantoprazole and esomeprazole) demonstrated a similar efficacy (Yeomans et al., 1998; Hawkey et al., 1998; Agrawal et al., 2000) in preventing NSAID-related bleeding (a mean reduction of at least four-fould).

Subsequently, in patients with previous GI complications, usually considered to be at exceptionally high risk of GI events, a combined treatment of a coxib with a PPI was proposed in order to reduce GI toxicity. This strategy was evaluated in a RCT comparing the combination therapy of celecoxib plus PPI to celecoxib plus placebo in *Hp* negative patients who presented an upper GI bleeding. The addiction of a PPI to a COX-2 selective inhibitor was demonstrated effective for prevention of ulcer re-bleeding (13-month incidence of 0% vs. 8.9% in patients treated with celecoxib alone) and considered the best treatment management in the very high risk group of patients (Chan et al., 2007).

#### *Patients requiring high-doses NSAIDs*

158 New Advances in the Basic and Clinical Gastroenterology

describe specific strategies for high-risk categories of patients; designed to significantly reduce side effects (both general and gastrointestinal) and the number needed to treat (NNT) to achieve the endpoints (reduction in GI events). These strategies often involve the prescription of a gastroprotecive agent (PPIs), reduction of nsNSAIDs dose and the change

Most guidelines do not suggest gastroprotective strategies for low-risk patients. Physicians prescribing NSAIDs to low-risk patients (less than 65 years, no comorbidity, no concomitant antiplatelet, anticoagulants or corticosteroids and no previous history of NSAID-related GI complications) should follow the general suggestions for GI complications reduction (eg.

However, even in patients without risk factors, two clinical trials demonstrated a significant reduction of GI events (detection of asymptomatic ulceration and bleeding) in patients receiving coxibs compared to those on nsNSAIDs (Bombardier et al., 2000; Silverstein et al., 2000); these evidences suggest that the low-risk category of patients could become a "no-

Epidemiological and retrospective studies identified a past episode of peptic ulcer as a risk factor for development of GI events in patients receiving chronic NSAID therapy; moreover, both *Hp* positive and negative patients present an increased relative risk of complications (Rockal et al., 1995). In order to reduce the risk of GI injury, the switch to a coxib was evaluated in patients with a past history of ulcer disease; both rofecoxib and celecoxib demonstrated their efficacy in reduction of GI events (from 8.8/100 patient/year with non selective NSAID to 2 with rofecoxib) (Laine, 2001). Therefore, in patients with a previous history of peptic ulcer disease, the switch to a selective COX-2 inhibitor could be considered

Although switching to a coxib induces a significant risk reduction, in these patients there is still a high residual risk of development of GI complications (10 events per 100 patients treated/year in VIGOR study) (Bombardier et al., 2000). In this setting, the combination therapy of a PPI to a standard NSAID appears to be more appropriate than a coxib alone; clinical studies demonstrated that combination with omeprazole induces a significant reduction in ulcer development both in patients with previous ulceration and perforation when compared to patients receiving a COX-2 selective inhibitors (rofecoxib) or monotherapy with ns-NSAID (ibuprofen) [Cullen et al., 1998; Hawkey et al., 1998). After this first evidence, PPIs other than omeprazole (both lansoprazole, pantoprazole and esomeprazole) demonstrated a similar efficacy (Yeomans et al., 1998; Hawkey et al., 1998; Agrawal et al., 2000) in preventing NSAID-related bleeding (a mean reduction of at least

Subsequently, in patients with previous GI complications, usually considered to be at exceptionally high risk of GI events, a combined treatment of a coxib with a PPI was proposed in order to reduce GI toxicity. This strategy was evaluated in a RCT comparing the combination therapy of celecoxib plus PPI to celecoxib plus placebo in *Hp* negative patients who presented an upper GI bleeding. The addiction of a PPI to a COX-2 selective inhibitor

prescribe the lowest effective dose and avoid drugs with high GI toxicity).

to selective COX-2 inhibitors. *Management of low-risk patients* 

risk" one, if well managed.

*Patient with history of peptic ulcer disease* 

a practical and cost-effective strategy.

four-fould).

When physicians have to prescribe high NSAID doses, there is a significant increase in the relative risk of development of GI complications (about three-fold).(Henry et al., 1996; Langman et al., 1994). In those cases, pharmacological and clinical evidences demonstrated that coxibs are safer than nsNSAIDs with a similar anti-inflammatory and analgesic effects. However, high-doses of coxibs show an overall increased risk of adverse events (both CV and related to fluid retention) and this should be taken into account in each single case to balance the risk/benefit ratio..

## *Helicobacter pylori positive*

Large population-based studies and meta-analysis demonstrated that *Hp* infection induce a two-fold increase in the risk of developing peptic complications in patients receiving NSAIDs (Chan et al., 2002; Vergara et al., 2005). Moreover, also in patients receiving coxibs, *H pylori* remain a risk factor for development of GI ulcers and bleeding (Bombardier et al., 2000). Systematic reviews and meta-analysis (Chan, 1997; Chan 2002; Vergara 2005) confirmed the efficacy of *Hp* eradication in preventing upper GI complications in patients on chronic NSAIDs therapy, even though treating *Helicobacter pylori* does not completely abolish the risk of bleeding in high risk patients (Chan, 2001). Therefore, even though it is often underestimated in general clinical practice, a test-and-treat strategy is mandatory in patients who require long-term NSAIDs therapy (Gabriel, 1991; Wolfe, 1999; Sauerbaum, 2002; Barkin, 1998; Laine, 2002; Chan, 2002; Laine, 1992; Hawkey, 1998; Loeb, 1992; Aalykke,*.*1999; Cullen, 1997).

#### *Patients on corticosteroids*

Corticosteroids have a synergistic effect with NSAID, magnifying their GI toxicity, and an intrinsic gastrolesive potential effect, specially in patients with multiple concurrent diseases; the risk of ulcer development is increased both in patients receiving NSAIDs and in non-NSAID users. A correct strategy, for the management of patients who need both corticosteroids and NSAIDs was suggested by post-hoc analysis of large clinical trials, showing that prescription of coxibs seems to reduce the risk of GI complications.

When prescribing NSAIDs to patients requiring high-doses of corticosteroids, a management strategy able to guarantee a reduced GI risk seems to be the choice of a coxib coupled by gastroprotection with a PPI; however, specific data in this setting are lacking (Holvoet et al., 1991; Hochain et al., 1995; Laine et al., 2010; Weil et al., 2000).

#### *Use of anticoagulant agents or patients affected by coagulopathy*

Co-prescription of anticoagulants and NSAIDs induce a significant increase in GI events and bleeding (both clinically manifest and occult); consequently, in these patients NSAID prescription should be considered a contraindication. Although there is a lack of specific data, in those patients when necessary, the co-prescription of a selective COX-2 inhibitor plus a PPI have to be considered in order to reduce the high risk of bleeding and the high morbidity and mortality that goes with it.

Chronic NSAIDs Therapy and Upper Gastrointestinal Tract – Mechanism of Injury,

receiving both NSAIDs and aspirin (Lai et al., 2002)

not reduce GI events (Chan et al., 2005).

*Gastroprotective strategies in patients receiving clopidogrel* 

PPI-receiving vs 0.43 of H2RA-receiving) (Ng et al., 2008).

gastroprotection.

2006).

strategy should be offered.

Mucosal Defense, Risk Factors for Complication Development and Clinical Management 161

 Aspirin induce a 2- to 4-fold increase in risk of development of GI injury with a dosedependent effect; therefore <80 mg/day should be preferred coupled to

When prescription of low dose of aspirin is associated to NSAIDs a gastroprotective

PPIs are demonstrated as the most effective gastroprotective agents in patients

 When aspirin or clopidogrel (or both) are prescribed together with anticoagulant agents (heparin, fractionated heparin or oral anticoagulant) a significant increase in upper GI bleeding risk is observed. Combination of antiplatelet and anticoagulant agent must be prescribed only with a clear indication (vascular, valvular or arrhythmic). In order to reduce the overall bleeding risk (both extracranial and intracranial) when warfarin is co-administered to aspirin, INR must be < 2.5 (Andreotti et al., 2006; Zhurram et al.,

 In high-risk patients, ASA and non-ASA antiplatelet agents (ticlopidine and clopidogrel) present similar bleeding risk, therefore, switching aspirin to clopidogrel do

Dual anti-platelet therapy (low-dose aspirin plus clopidogrel), prescribed to patients for secondary prevention of acute coronary syndrome or undergoing coronary stent implantation, is effective in preventing stent thrombosis and reducing the risk of reinfarction, but significantly increase the risk of GI bleeding. The relative risk increase to about 2.5-fold in patients receiving clopidogrel or ASA, when compared to patients not on antiplatelet agents (Ibanez et al., 2006; van Hecken et al., 1998; Delaney et al., 2007). Use of clopidogrel in aspirin-taking patients synergistically increase the risk of bleeding (2- to 3 fold) and the mean blood loss in case of haemorrhage (Yusuf et al., 2001; Connoly et al., 2009). Dual antiplatelet agents are not indicated for CV primary prevention because of the

observed low reduction in CV events and significant increase in severe GI bleeding.

thereby reducing the rate of complications from upper GI injury (Ma et al., 2001).

Clopidogrel, as discussed above, does not induce ulceration of upper GI tract, but impairs natural healing process (through inhibition of platelet attivation and aggregation) and increases bleeding from preexisting lesions (induced by other causes). As in NSAID-related GI bleeding, acid suppression could favors the healing process and stabilization of thrombi

Acid suppressive therapy (both H2RAs and PPIs) demonstrated its efficacy in reduction of bleeding risk related to antiplatelet therapy. H2RAs appear to be able to reduce the rate of GI adverse events in patients receiving low-dose aspirin (3.8% in famotidine-receivers vs. 23.5% of placebo ones) (Taha et al., 2009) while no reduction was found in those treated with clopidogrel (Lanas et al., 2007). PPIs resulted to be more effective than H2RAs in reducing upper GI events in a cohort of patients receiving both aspirin and clopidogrel (OR:0.04 of

After the evidence of the positive effects of PPI prescription in reducing GI adverse events among clopidogrel-receiving patients, some observational studies suggested the presence of a possible interaction between clopidogrel and PPI determining a reduced antiplatelet effect (Ho et al., 2009; Juurlink et al., 2009). Moreover, in vitro studies (assessing platelet
