Provision of Gastroprotective Medication and Bleeding Risk Following Acute Coronary Syndrome

Abstract: Background. Gastrointestinal (GI) bleeding following percutaneous coronary intervention (PCI) is associated with increased mortality. ACCF/AHA/SCAI guidelines recommend prophylaxis to prevent GI bleeding in patients, with the highest GI bleeding risks taking dual-antiplatelet therapy (DAPT). The REPLACE risk score identifies factors predictive of peri-PCI bleeding from vascular access and non-access sites. We determined whether high bleeding risk acute coronary syndrome (ACS) patients taking DAPT were appropriately provided with GI prophylaxis and investigated the association between age and clinical presentation on the likelihood of receiving prophylactic therapy. Methods. This is a retrospective analysis of all non-elective PCI patients at a single center between May and December 2008 stratified by age (<65, 65-74, and ≥75 years). REPLACE scores were calculated and discharge medication was obtained from case records. Results. Complete discharge medication data were available for 800 patients (median age, 63 years; 45.1% with ST-elevation myocardial infarction [STEMI]). A total of 370 patients (46.3%) were high bleeding risk (REPLACE scores ≥10), including all patients ≥75 years (n = 173), 83.5% of patients 65-74 years (n = 177), and 4.8%  of patients <65 years (n = 20). In total, 97.6% were discharged on DAPT. Within the high bleeding risk group, 45.1% received GI prophylaxis. Patients 65-74 years were least likely to receive prophylaxis (<65 years, 60%; 65-74 years, 38.4%; ≥75 years, 50.3%; P<.03). Presentation with STEMI was independently associated with a reduced likelihood of GI prophylaxis provision (odds ratio, 0.63; 95% confidence interval, 0.40-0.99; P=.045). Conclusions. Less than half of ACS patients at high bleeding risk taking DAPT are provided with GI prophylaxis. Increased use of objective bleeding risk scores may help guide risk/benefit decisions in patients taking clopidogrel and proton pump inhibitors.

J INVASIVE CARDIOL 2013;25(8):397-401

Key words: acute coronary syndrome, risk stratification, bleeding, delivery of care

______________________________

Percutaneous coronary intervention (PCI) for acute coronary syndrome (ACS) is increasingly undertaken in elderly patients with multiple comorbidities and high bleeding risk. Peri-PCI bleeding is a major predictor of adverse outcome and is associated with increased mortality even after adjustment for confounding factors.^1,2 Gastrointestinal (GI) bleeding following PCI occurs in 1.0%-2.7% of patients and is associated with significantly higher in-hospital, 30-day, and 1-year mortality.^3-8 Increasing age is a strong predictor of peri-PCI GI bleeding with an incidence 2.79 times greater in patients >70 years.⁴ Patients taking dual-antiplatelet therapy (DAPT) are at higher risk of GI bleeding, although registry, case control, and randomized trial data suggest that this risk is significantly reduced by proton pump inhibitors (PPIs).^5,9,10 ACCF/ACG/AHA guidelines from October 2008¹¹ initially recommended that all patients on DAPT or with risk factors for GI bleeding should receive GI prophylaxis with PPIs, though randomized trial data to support this approach were lacking. Subsequent focused updates of ACCF/AHA/SCAI guidelines in 2011¹² continue to recommend PPIs for patients taking DAPT who have a prior history of GI bleeding (class I, level of evidence C) and are appropriate for patients who have an increased risk of GI bleeding (including those of advanced age, or taking steroids, warfarin or non-steroidal anti-inflammatories; class IIa, level of evidence C). PPIs are not recommended for routine use in patients at low risk of GI bleeding.

Using a validated peri-PCI bleeding risk score, we determined whether invasively managed ACS patients with the highest bleeding risk who were taking DAPT were appropriately prescribed GI prophylaxis to minimize their risk of GI bleeding. We also determined the relationship between age and mode of ACS presentation on the prescription of GI prophylaxis by physicians following PCI.

Methods

A single-center retrospective analysis of all PCI patients with ACS admitted to a tertiary referral center in the United Kingdom (Freeman Hospital, Newcastle-upon-Tyne) between May and December 2008 was performed. These dates were selected to precede the possible confounding influence upon prescribing practice of recommendations made in July 2009 by the Medicines and Healthcare Regulatory Agency (MHRA) advising against clopidogrel and PPI coadministration.¹³ REPLACE bleeding risk score¹⁴ (Figure 1) was calculated for all patients using data from the Myocardial Ischaemia National Audit (MINAP) database. The REPLACE score was chosen because it is well validated, utilizes only parameters available prior to angiography, and is predictive of both vascular access site and non-access site bleeding complications. It is thus easy to apply in clinical practice and relevant for physicians when selecting patients for invasive management.

Case notes and discharge medication prescriptions were reviewed for all patients. Patients who died during their index admission or in whom discharge medication data were incomplete were excluded from the analysis (Figure 2). Data were collected on antiplatelet, anticoagulant and GI prophylaxis (PPI or H2 antagonist) provision at the time of hospital discharge along with other guideline-recommended secondary prevention medications. Periprocedural bleeding events were identified by comparison of preprocedural and nadir postprocedural hemogolobin values and from case records, radiology, and endoscopy databases. The study population was stratified by age (<65 years, 65 to 74 years [elderly], and ≥75 years [very elderly]) and a high bleeding risk subgroup was identified using the REPLACE score (REPLACE score ≥10, equivalent to at least a 5% risk of major bleeding).¹⁴

Statistical analysis. Categorical variables are expressed as percentages. Continuous variables are presented as medians and interquartile ranges (IQRs). Group comparisons were made using the Kruskal-Wallis test for continuous variables. Chi-square test for trend was used to test for ordered categorical variables and Pearson’s chi-square to assess differences in medication provision. To evaluate the independent relation between age and the provision of gastric protection therapy, we adjusted for other prognosticators and confounders in multivariable logistic regression analysis. We considered age, gender, presence of diabetes, renal function, presentation with ST-elevation myocardial infarction (STEMI), discharge on aspirin, clopidogrel or warfarin, and post-PCI fall in hemoglobin (Hb) as candidate predictor variables^15,16 and developed a parsimonious model by backward elimination (for P<.05). All analyses were performed using SPSS version 15. A 2-sided P-value <.05 was considered statistically significant. 

Results

Entire study population. A total of 910 patients underwent non-elective PCI for ACS during the study period. Thirty-six patients (3.9%) died prior to discharge and complete discharge medication data were available for 800 of the remaining patients (91.5%). Of these, 415 (51.9%) were aged <65 years, 212 (26.5%) were between 65-74 years, and 173 (21.6%) were ≥75 years. Patient demographics and medications for the entire study population are presented in Table 1. Rates of adherence to secondary preventative therapies were high across all age groups. Female gender, cerebrovascular disease, peripheral vascular disease, and hypertension were more prevalent in older age groups, while serum creatinine was higher and hematocrit was lower across the increasing age groups (all P<.001). 

High bleeding risk subgroup. Three hundred and seventy patients (46.3%) had REPLACE scores of ≥10. These included all patients ≥75 years (n = 173), 83.5% of those 65-74 years (n = 177), and 4.8% of those <65 years (n = 20). Patient demographics and medication provision for the high bleeding risk subgroup are shown in Table 2. Consistent with the factors used to identify bleeding risk in the REPLACE score algorithm, lower hematocrit and higher serum creatinine were more prevalent in younger patients, together with hypercholesterolemia and diabetes, while only cerebrovascular disease remained more common in elderly patients.

Discharge medication in high bleeding risk subgroup. At the time of discharge, a total of 347 high bleeding risk patients (93.8%) were taking both aspirin and clopidogrel. Thirteen patients (3.5%) were prescribed triple therapy (aspirin, clopidogrel, and warfarin), 4 (1.1%) aspirin monotherapy, 5 (1.4%) clopidogrel monotherapy, and 1 (0.3%) received aspirin and warfarin. There were no differences in the rate of prescription of DAPT between age groups. With the exception of statins, prescription of other guideline-recommended ACS secondary preventative medications did not differ between age groups. Antiplatelet, anticoagulant, and GI prophylaxis prescriptions stratified by age group are illustrated in Figure 3. Less than half of the high bleeding risk cohort received GI prophylaxis. Despite equivalent eligibility (REPLACE scores ≥10), GI prophylaxis provision differed between age groups (P=.03), with the lowest rates of provision in patients 65-74 years. There was a numerical trend toward lower provision of GI prophylaxis in all patients >65 years (44.3% vs 60%) that did not reach statistical significance (P=.17).

Multivariable analysis of high bleeding risk subgroup. In the high bleeding risk subgroup, presentation with STEMI was independently associated with a reduced likelihood of GI prophylaxis (odds ratio, 0.63; 95% confidence interval, 0.40-0.99; P=.045). While overall age was an independent negative predictor, this relation was not apparent when assessed according to the prespecified subgroups of ages 65-74 years or ≥75 years. Because there were no patients who met the combined criteria of age <65 years, high bleeding risk, and STEMI, our model was unable to discriminate the individual influence of age ≥65 years and presentation with STEMI on provision of GI prophylaxis.

In-hospital bleeding events and hemoglobin levels in the high bleeding risk subgroup. Post-PCI falls in Hb within each age group are illustrated in Figure 4. A total of 189 patients (51.1%) experienced a minor fall in Hb (<3 g), 13 patients (3.5%) had Hb falls of 3-5 g, and 1 patient (0.3%) from age group 65-74 years had an Hb fall of >5 g. There were no significant differences between age groups in the frequency of any bleeding event or bleeding events of >3 g.

Discussion

The main findings of this study in patients undergoing PCI for ACS are: (1) the majority of high bleeding risk patients are not prescribed GI prophylaxis while taking DAPT; (2) most elderly and all very elderly patients were at high bleeding risk, yet were less likely to be prescribed GI prophylaxis than younger patients with high bleeding risk; and (3) presentation with STEMI was independently associated with a decreased likelihood of GI prophylaxis prescription.

Prescribing PPIs for the primary prevention of GI bleeding following PCI remains controversial. In 2009, the MHRA advised against the coprescription of PPIs with clopidogrel¹⁷ based on a number of pharmacokinetic studies demonstrating that omeprazole reduced the platelet inhibitory activity of clopidogrel.^18,19 Subsequent registry data have shown an association with increased risk of myocardial infarction²⁰ and death²¹ in patients taking this combination. However, data from the randomized prospective COGENT study⁹ and post hoc analysis of the TRITON-TIMI-38 and PRINCIPLE-TIMI-44 trials²² have questioned the clinical relevance of this interaction and shown that PPIs reduce major bleeding without an increase in ischemic events. A recent focused update of the ACCF/ACG/AHA¹⁷ and ACCF/AHA/SCAI¹² guidelines recommends careful consideration of the risks and benefits of co-prescription of PPIs with clopidogrel, but continues to recommend GI prophylaxis with PPIs for patients at high bleeding risk, particularly in those with previous GI bleeds and the elderly. MHRA advice has also been updated.²³

We found that all of our patients ≥75 years and the majority of patients >65 years had REPLACE scores ≥10, suggesting that use of the REPLACE score to identify high bleeding risk had little additive value compared to age alone. In contrast, patients <65 years were almost exclusively at low to moderate risk, with less than 5% of young patients in the highest bleeding risk category. Use of objective scoring systems may thus be a more useful guide to the relative risks and benefits of PPIs in younger patients.

Despite the well-recognized influence of age on bleeding risk, our analysis revealed that elderly and very elderly patients were less likely to receive GI prophylaxis than their younger counterparts. This paradox is particularly striking when we consider that the absolute REPLACE score within the high bleeding risk cohort was higher in these elderly groups (mean REPLACE score was 18 in patients ≥75 years vs 11 in patients <65 years) and that elderly patients more commonly received triple therapy (DAPT + warfarin). The reasons underlying this risk-treatment paradox in the provision of GI prophylaxis are not immediately apparent. We^24,25 and others²⁶ have previously reported a similar risk-treatment paradox in ACS patients with respect to the provision of guideline-recommended medical therapy or referral for coronary angiography. In these instances, there was a definite possibility of harm with the recommended treatment — for example, an increased risk of bleeding with guideline-recommended antiplatelet/anticoagulant therapy or the procedural risks associated with invasive ACS management. At the time of this study, there were no published concerns regarding use of PPIs. Hence, we cannot conclude that physicians failed to provide PPIs to elderly patients because of safety concerns. Although not tested in this study, we postulate that it was a subjective underestimation of the importance of age in determining bleeding risk that accounted for the underprovision of PPIs in the elderly.

Study limitations. This was a single-center retrospective registry study and similar prescription practices may not been seen in other hospitals. However, the Freeman Hospital is a tertiary referral center with all cardiology subspecialties represented and may be considered representative of the broad spectrum of cardiology practice in the United Kingdom. Complete discharge medication data were not available for 74 patients (9.2%). These were cases where the patient was transferred back to their referring hospital prior to discharge. We cannot exclude that the pattern of discharge provision of PPIs might have differed in these hospitals. 

The REPLACE score has a number of limitations for guiding GI prophylaxis provision. First, it does not include data on previous GI bleeding, perhaps the strongest predictor of GI bleeding risk. Second, in common with other peri-PCI bleeding risk estimators,^{27, 28} predictive data were derived from a cohort of patients using femoral rather than radial access. The majority of patients in our cohort underwent radial procedures. However, two-thirds of peri-PCI bleeds are non-access site related,¹ and our study was designed to assess uptake of strategies to prospectively reduce non-access site bleeds following hospital discharge. Third, bleeding risk estimates were derived from in-hospital bleeding events and their predictive value for future GI bleeds is unproven. Conversely, prospective trials in other patient populations²⁹ have identified age >65 years, previous GI bleeding, and concurrent aspirin, steroid, or anticoagulant use as the strongest predictors of GI bleeding. The majority of ACS patients will remain on DAPT for up to 1 year and it is likely that those patients identified as having high periprocedural bleeding risk will remain at high bleeding risk. 

Conclusion

Patients with high bleeding risk taking DAPT were not routinely provided with GI prophylaxis. Increased use of objective bleeding risk scores may help guide risk/benefit decisions in patients taking clopidogrel who are considered for PPIs. Further research on the safety and efficacy of PPIs for the primary prevention of GI bleeding in patients taking DAPT is warranted.

Acknowledgment. We are grateful to Mrs Sheila Jamieson for her assistance in the preparation of this manuscript.

References

1. Verheugt FW, Steinhubl SR, Hamon M, et al. Incidence, prognostic impact, and influence of antithrombotic therapy on access and nonaccess site bleeding in percutaneous coronary intervention. JACC Cardiovasc Interv. 2011;4(2):191-197.
2. Rao SV, O’Grady K, Pieper KS, et al. Impact of bleeding severity on clinical outcomes among patients with acute coronary syndromes. Am J Cardiol. 2005;96(9):1200-1206.
3. Nikolsky E, Mehran R, Sadeghi HM, et al. Prognostic impact of blood transfusion after primary angioplasty for acute myocardial infarction: analysis from the CADILLAC (Controlled Abciximab and Device Investigation to Lower Late Angioplasty Complications) trial. JACC Cardiovasc Interv. 2009;2(7):624-632.
4. Abbas AE, Brodie B, Dixon S, et al. Incidence and prognostic impact of gastrointestinal bleeding after percutaneous coronary intervention for acute myocardial infarction. Am J Cardiol. 2005;96(2):173-176.
5. Chin MW, Yong G, Bulsara MK, et al. Predictive and protective factors associated with upper gastrointestinal bleeding after percutaneous coronary intervention: a case-control study. Am J Gastroenterol. 2007;102(11):2411-2416.
6. Alli O, Smith C, Hoffman M, et al. Incidence, predictors, and outcomes of gastrointestinal bleeding in patients on dual antiplatelet therapy with aspirin and clopidogrel. J Clin Gastroenterol. 2011;45(5):410-414.
7. Shivaraju A, Patel V, Fonarow GC, et al. Temporal trends in gastrointestinal bleeding associated with percutaneous coronary intervention: analysis of the 1998-2006 Nationwide Inpatient Sample (NIS) database. Am Heart J. 2011;162(6):1062-1068.e5.
8. Gaglia MA Jr, Torguson R, Gonzalez MA, et al. Correlates and consequences of gastrointestinal bleeding complicating percutaneous coronary intervention. Am J Cardiol. 2010;106(8):1069-1074.
9. Bhatt DL, Cryer BL, Contant CF, et al. Clopidogrel with or without omeprazole in coronary artery disease. N Engl J Med. 2010;363(20):1909-1917.
10. Ng FH, Wong SY, Lam KF, et al. Gastrointestinal bleeding in patients receiving a combination of aspirin, clopidogrel, and enoxaparin in acute coronary syndrome. Am J Gastroenterol. 2008;103(4):865-871.
11. Bhatt DL, Scheiman J, Abraham NS, et al. ACCF/ACG/AHA 2008 expert consensus document on reducing the gastrointestinal risks of antiplatelet therapy and NSAID use: a report of the American College of Cardiology Foundation Task Force on Clinical Expert Consensus Documents. J Am Coll Cardiol. 2008;52(18):1502-1517.
12. Levine GN, Bates ER, Blankenship JC, et al. 2011 ACCF/AHA/SCAI Guideline for Percutaneous Coronary Intervention: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines and the Society for Cardiovascular Angiography and Interventions. Circulation. 2011;124(23):e574-e651.
13. Medicines and Healthcare Products Regulatory Agency website. https://www.mhra.gov.uk/Safetyinformation/DrugSafetyUpdate/CON087962. Accessed Dec, 2012.
14. Nikolsky E, Mehran R, Dangas G, et al. Development and validation of a prognostic risk score for major bleeding in patients undergoing percutaneous coronary intervention via the femoral approach. Eur Heart J. 2007;28(16):1936-1945.
15. Moscucci M, Fox KA, Cannon CP, et al. Predictors of major bleeding in acute coronary syndromes: the Global Registry of Acute Coronary Events (GRACE). Eur Heart J. 2003;24(20):1815-1823.
16. Kinnaird TD, Stabile E, Mintz GS, et al. Incidence, predictors, and prognostic implications of bleeding and blood transfusion following percutaneous coronary interventions. Am J Cardiol. 2003;92(8):930-935.
17. Abraham NS, Hlatky MA, Antman EM, et al. ACCF/ACG/AHA 2010 expert consensus document on the concomitant use of proton pump inhibitors and thienopyridines: a focused update of the ACCF/ACG/AHA 2008 expert consensus document on reducing the gastrointestinal risks of antiplatelet therapy and NSAID use. A report of the American College of Cardiology Foundation Task Force on Expert Consensus Documents. J Am Coll Cardiol. 2010;56(24):2051-2066.
18. Gilard M, Arnaud B, Le Gal G, et al. Influence of omeprazol on the antiplatelet action of clopidogrel associated to aspirin. J Thromb Haemost. 2006;4(11):2508-2509.
19. Gilard M, Arnaud B, Cornily JC, et al. Influence of omeprazole on the antiplatelet action of clopidogrel associated with aspirin: the randomized, double-blind OCLA (Omeprazole CLopidogrel Aspirin) study. J Am Coll Cardiol. 2008;51(3):256-260.
20. Stockl KM, Le L, Zakharyan A, et al. Risk of rehospitalization for patients using clopidogrel with a proton pump inhibitor. Arch Intern Med. 2010;170(8):704-710.
21. Ho PM, Maddox TM, Wang L, et al. Risk of adverse outcomes associated with concomitant use of clopidogrel and proton pump inhibitors following acute coronary syndrome. JAMA. 2009;301(9):937-944.
22. O’Donoghue ML, Braunwald E, Antman EM, et al. Pharmacodynamic effect and clinical efficacy of clopidogrel and prasugrel with or without a proton-pump inhibitor: an analysis of two randomised trials. Lancet. 2009;374(9694):989-997.
23. Medicines and Healthcare Products Regulatory Agency website. https://www.mhra.gov.uk/Safetyinformation/DrugSafetyUpdate/CON087711#ref2. Accessed Dec, 2012.
24. Bagnall AJ, Goodman SG, Fox KA, et al. Influence of age on use of cardiac catheterization and associated outcomes in patients with non-ST elevation acute coronary syndromes. Am J Cardiol. 2009;103(11):1530-1536.
25. Bagnall AJ, Yan AT, Yan RT, et al. Optimal medical therapy for non-ST segment elevation acute coronary syndromes: exploring why physicians do not prescribe evidence-based treatment and why patients discontinue medications after discharge. Circ Cardiovasc Qual Outcomes. 2010;3(5):530-537.
26. Fox KA, Anderson FA Jr, Dabbous OH, et al. Intervention in acute coronary syndromes: do patients undergo intervention on the basis of their risk characteristics? The Global Registry of Acute Coronary Events (GRACE). Heart. 2007;93(2):177-182.
27. Mehran R, Pocock SJ, Nikolsky E, et al. A risk score to predict bleeding in patients with acute coronary syndromes. J Am Coll Cardiol. 2010;55(23):2556-2566.
28. Subherwal S, Bach RG, Chen AY, et al. Baseline risk of major bleeding in non-ST segment-elevation myocardial infarction: the CRUSADE (Can Rapid risk stratification of Unstable angina patients Suppress ADverse outcomes with Early implementation of the ACC/AHA Guidelines) Bleeding Score. Circulation. 2009;119(14):1873-1882.
29. Laine L, Curtis SP, Cryer B, et al. Risk factors for NSAID-associated upper GI clinical events in a long-term prospective study of 34,701 arthritis patients. Aliment Pharmacol Ther. 2010;32(10):1240-1248.