Abstract
Background
Safety of proton pump inhibitors (PPIs) in patients undergoing percutaneous coronary intervention (PCI) with aspirin-free P2Y12 inhibitor monotherapy was unknown.
Objectives
The authors aimed to evaluate effects of PPIs in patients undergoing PCI with P2Y12 inhibitor monotherapy.
Methods
We compared outcomes between patients with and without PPI prescription in the STOPDAPT-3 trial enrolling patients undergoing PCI stratified by the no-aspirin (1-month prasugrel monotherapy followed by clopidogrel monotherapy: n = 2,909 [acute coronary syndrome: n = 2,170, high bleeding risk: n = 1,580]) and the aspirin (1-month dual antiplatelet therapy followed by aspirin monotherapy: n = 2,914 [acute coronary syndrome: n = 2,171, high bleeding risk: n = 1,566]) strategies at 1 year.
Results
PPIs were prescribed in 2,418 patients (83.1%) with the no-aspirin strategy, and in 2,695 patients (92.5%) with the aspirin strategy. In the propensity score matched cohort (no-aspirin strategy: n = 902 and aspirin strategy: n = 376), a composite of cardiovascular death, myocardial infarction, definite stent thrombosis, or stroke more often occurred in the PPI group than in the no-PPI group with the no-aspirin strategy (7.1% vs 2.4%, P = 0.002), but not with the aspirin strategy (6.9% vs 7.4%, P = 0.817). Death also more often occurred in the PPI group than in the no-PPI group with the no-aspirin strategy, but not with the aspirin strategy. Incidence of major bleeding was not different between the groups regardless of the no-aspirin and aspirin strategies (5.5% vs 3.3%, P = 0.150, and 6.9% vs 4.3%, P = 0.278)
Conclusions
PPI use was associated with higher risks of cardiovascular events and mortality without decreasing major bleeding in patients undergoing PCI with aspirin-free P2Y12 inhibitor monotherapy. (ShorT and OPtimal duration of Dual AntiPlatelet Therapy after everolimus-eluting cobalt-chromium stent-3 [STOPDAPT-3]; NCT04609111)
Key Words: antiplatelet therapy, coronary stent, percutaneous coronary intervention, PPI
Central Illustration
Gastrointestinal bleeding was a common complication in patients who underwent percutaneous coronary intervention (PCI) because dual antiplatelet therapy (DAPT) with P2Y12 inhibitors and aspirin was mandatory to prevent cardiovascular events after PCI.1,2 Previous randomized clinical trials demonstrated that proton pump inhibitors (PPIs) reduced upper gastrointestinal bleeding in patients receiving DAPT or aspirin monotherapy.3, 4, 5 Recently, P2Y12 inhibitor monotherapy after very short DAPT has become a preferred strategy in real clinical practice based on the results of the several randomized clinical trials demonstrating that P2Y12 inhibitor monotherapy after very short DAPT compared with standard DAPT reduced major bleeding without increasing cardiovascular events after PCI.6, 7, 8 There were few data on the efficacy of PPIs in patients receiving P2Y12 inhibitor monotherapy. In addition, previous studies suggested that PPIs might reduce antiplatelet effect of P2Y12 inhibitors, especially in clopidogrel.9, 10, 11, 12 However, data on the safety of PPIs in patients receiving P2Y12 inhibitor monotherapy were scarce. Recently, we conducted the STOPDAPT-3 (ShorT and Optimal Duration of Dual AntiPlatelet Therapy-3) trial that compared aspirin-free P2Y12 inhibitor monotherapy (1-month prasugrel monotherapy followed by clopidogrel monotherapy) with 1-month DAPT followed by aspirin monotherapy after PCI.13,14 In the present study, we evaluated the efficacy and safety of PPIs in patients with an aspirin-free P2Y12 inhibitor monotherapy and in patients with an aspirin-based strategy using the dataset of the STOPDAPT-3 trial.
Methods
Study design and population
The STOPDAPT-3 (NCT04609111) was a physician-initiated, prospective, multicenter, open-label, adjudicator-blinded randomized clinical trial, where we compared the group of 1-month prasugrel monotherapy followed by clopidogrel monotherapy with the group of 1-month DAPT followed by aspirin monotherapy in patients planned for PCI in terms of cardiovascular and bleeding endpoints. The details for the study design and the overall results at 30 days and at 1 year were previously reported.13,14 Briefly, the trial enrolled patients with acute coronary syndrome or those with high bleeding risk by the criteria of the Academic Research Consortium for High Bleeding Risk irrespective of acute coronary syndrome who were planned for PCI with cobalt-chromium everolimus-eluting stents (Xience series, Abbott Vascular). Patients were randomly assigned in a 1-to-1 fashion to the 1-month aspirin-free prasugrel monotherapy or the 1-month DAPT with aspirin and prasugrel. At 1 month (between 30 and 59 days after the index PCI), patients who had received prasugrel monotherapy were switched to clopidogrel (75 mg/d) monotherapy (no-aspirin strategy), and patients who had received DAPT were switched to aspirin (81-100 mg/d) monotherapy (aspirin strategy). Each group of patients continued the assigned antiplatelet monotherapy up to 1 year. The ethical committees in all the participating centers approved the study protocol, and informed consent was obtained from all patients.
We evaluated prescription for PPIs at the time of discharge from the index hospitalization for PCI. Vonoprazan was included in the PPIs. In the present study, we compared clinical outcomes between patients with and without PPI prescription stratified by the no-aspirin and aspirin strategies. After the exclusion of 3 patients who did not receive PCI for the absence of suitable coronary lesions, 37 patients who withdrew consent, and 139 patients who died during index hospitalization, the study population consisted of 2,909 patients with the no-aspirin strategy and 2,914 patients with the aspirin strategy (Figure 1). The prevalence of acute coronary syndrome and Academic Research Consortium for High Bleeding Risk was 2,170 patients (74.6%) and 1,580 patients (54.3%) in 2,909 patients with the no-aspirin strategy, and 2,171 patients (74.5%) and 1,566 patients (53.7%) in 2,914 patients with the aspirin strategy. We excluded patients who died during index hospitalization because PPI prescription was evaluated only at discharge, and many of the patients who died during index hospitalization were unlikely to take oral agents because of their serious clinical conditions. On the other hand, we did not exclude patients who experienced nonfatal clinical events during the index hospitalization, and the follow-up in the present study commenced on the day of randomization, because the vast majority of the patients with PPIs at discharge were likely to have taken PPIs on or before the day of randomization.
Figure 1.
Study Flow
The study population consisted of 2,909 patients with the no-aspirin strategy and 2,914 patients with the aspirin strategy. After the propensity score matching, there were 451 patients each in the PPI and no-PPI groups with the no-aspirin strategy, 188 patients each in the PPI and no-PPI groups with the aspirin strategy. DAPT = dual antiplatelet therapy; PCI = percutaneous coronary intervention; PPI = proton pump inhibitor; STOPDAPT-3 = ShorT and OPtimal duration of Dual AntiPlatelet Therapy after everolimus-eluting cobalt-chromium stent-3.
Endpoints
The co-primary cardiovascular endpoint was a composite of cardiovascular death, myocardial infarction, definite stent thrombosis, or ischemic stroke, and the co-primary bleeding endpoint was major bleeding defined as the Bleeding Academic Research Consortium (BARC) type 3 or 5.15 The major secondary endpoint was a composite of cardiovascular death, myocardial infarction, definite stent thrombosis, ischemic stroke, or major bleeding (BARC 3 or 5), which represented a net adverse clinical outcome for cardiovascular and bleeding events. Myocardial infarction and stent thrombosis were defined by the Academic Research Consortium criteria.16 The definitions of other secondary endpoints are described in the supplemental materials. The independent clinical event committee adjudicated all the clinical events in a blinded fashion to the assigned group.
Statistical analysis
Categorical variables were presented as number and percentage and were compared using the chi-square test. Continuous variables were expressed as mean ± SD or median (IQR) and were compared using the Student’s t-test or Wilcoxon rank-sum test depending on their distributions. Cumulative incidences were estimated by the Kaplan-Meier methods, and differences were assessed with the log-rank test. We performed propensity score matching analyses as the main analyses to adjust the potential confounders stratified by the no-aspirin and aspirin strategies. A logistic regression model was used to develop propensity score with 22 clinically independent variables relevant to the choice of prescription for PPI (age ≥75 years, men, body mass index <25 kg/m2, acute coronary syndrome, current heart failure, prior myocardial infarction, prior stroke, prior heart failure, atrial fibrillation, peripheral artery disease, hypertension, diabetes, current smoker, hemoglobin <11.0 g/dL, severe chronic kidney disease [estimated glomerular filtration rate <30 mL/min/1.73 m2 or hemodialysis], long-term use of oral nonsteroidal anti-inflammatory drugs or steroids, staged PCI, renin-angiotensin system inhibitors, mineralocorticoid receptor antagonists, beta-blockers, statins, and histaminn H2 receptor blockers). Patients in the PPI group were matched to those in the no-PPI group using 1:1 greedy matching technique to create the propensity score matched cohort. Because PPI prescription was evaluated only at discharge, we excluded patients who died during index hospitalization before developing propensity score and creating the propensity score matched cohort. The effects of the PPI group relative to the no-PPI group were expressed as HRs and their 95% CIs estimated by the Cox proportional hazard models. Proportional hazard assumptions were assessed on the plots of log (times) vs log (−log[survival]) stratified by variables, and were verified to be acceptable. As the sensitivity analyses, we constructed the multivariable Cox proportional hazard models with the potential confounders as the risk-adjusting variables in the entire study population. The detailed methods for the multivariable analyses are described in the supplemental material. In addition, we conducted the sensitivity analyses in which the time of the discharge was regarded as time 0 in the propensity score matched cohort.
All reported P values were 2-sided. P values <0.05 were considered statistically significant. All analysis was performed with JMP pro version 17.0 (SAS Institute Inc). Details of the study organization, participating centers, endpoint definitions, and supplemental methods are provided in the Supplemental Appendix.
Results
Baseline characteristics
In the entire study population, PPIs were prescribed at discharge in 2,418 patients (83.1%) with the no-aspirin strategy and in 2,695 patients (92.5%) with the aspirin strategy (Figure 1). Lansoprazole and vonoprazan were commonly prescribed PPIs with the no-aspirin strategy (36.8% and 35.2%) and with the aspirin strategy (36.4% and 38.0%) (Supplemental Table 1).
In patients with the no-aspirin strategy, patients in the PPI group more often presented as acute coronary syndrome, cardiogenic shock, and current heart failure than those without (Supplemental Table 1). Patients in the PPI group more often had low body weight and comorbidities such as history of heart failure and left ventricular systolic dysfunction than those without. Patients in the PPI group less often had atrial fibrillation and hypertension compared with those without. The prescription rates of renin-angiotensin system inhibitors, mineralocorticoid receptor antagonists, beta-blockers, and statins were higer in the PPI group than in the no-PPI group, whereas the prescription rates of histamine H2 receptor blockers were lower in the PPI group than in the no-PPI group.
In patients with the aspirin strategy, patients in the PPI group were younger and more often presented as acute coronary syndrome than those in the no-PPI group (Supplemental Table 1). Patients in the PPI group less often had comorbidities such as prior myocardial infarction, peripheral artery disease, chronic kidney disease, and long-term use of oral nonsteroidal anti-inflammatory drugs or steroids than those in the no-PPI group. The prescription rate of renin-angiotensin system inhibitors, mineralocorticoid receptor antagonists, beta-blockers, and statins was higher in the PPI group than in the no-PPI group, whereas the prescription rate of histamine H2 receptor blockers was lower in the PPI group than in the no-PPI group.
After the propensity score matching, there were 902 patients (451 patients each in the PPI and no-PPI groups) with the no-aspirin strategy, and 376 patients (188 patients each in the PPI and no-PPI groups) with the aspirin strategy (Figure 1). In the propensity score matched cohort, baseline characteristics were well balanced between the PPI and no-PPI groups (Table 1). The prescription rate of antiplatelet agents with the aspirin strategy was higher in the PPI group than in the no-PPI group in the propensity score matched cohort.
Table 1.
Baseline Characteristics in the Propensity Matched Cohort
| No-Aspirin Strategy |
Aspirin Strategy |
|||||
|---|---|---|---|---|---|---|
| PPI Group | No-PPI Group | SMD | PPI Group | No-PPI Group | SMD | |
| (n = 451) | (n = 451) | (n = 188) | (n = 188) | |||
| Type of PPI | ||||||
| Omeprazole | 10 (2.2) | 0 (0) | 2 (1.1) | 0 (0) | ||
| Lansoprazole | 162 (36.0) | 0 (0) | 65 (34.6) | 0 (0) | ||
| Rabeprazole | 50 (11.1) | 0 (0) | 19 (10.1) | 0 (0) | ||
| Esomeprazole | 69 (15.3) | 0 (0) | 33 (17.6) | 0 (0) | ||
| Vonoprazan | 160 (35.5) | 0 (0) | 69 (36.7) | 0 (0) | ||
| Patient demographics | ||||||
| Age, y | 71.6 ± 11.6 | 71.7 ± 11.5 | 0.010 | 73.8 ± 11.7 | 73.9 ± 9.8 | 0.013 |
| ≥75 | 204 (45.2) | 215 (47.7) | 0.049 | 103 (54.8) | 101 (53.7) | 0.021 |
| Men | 347 (76.9) | 346 (76.7) | 0.005 | 132 (70.2) | 143 (76.1) | 0.132 |
| Body mass index, kg/m2 | 24.2 ± 3.8 | 24.3 ± 4.1 | 0.010 | 22.9 ± 3.4 | 23.7 ± 4.8 | 0.204 |
| <25 | 273 (60.5) | 288 (63.9) | 0.069 | 142 (75.5) | 130 (69.1) | 0.143 |
| Clinical presentation | ||||||
| Acute coronary syndrome | 305 (67.6) | 296 (65.6) | 0.042 | 104 (55.3) | 99 (52.7) | 0.053 |
| ST-segment elevation myocardial infarction | 149 (33.0) | 122 (27.1) | 54 (28.7) | 38 (20.2) | ||
| Non-ST-segment elevation myocardial infarction | 82 (18.2) | 77 (17.1) | 26 (13.8) | 25 (13.3) | ||
| Unstable angina | 74 (16.4) | 97 (21.5) | 24 (12.8) | 36 (19.1) | ||
| Cardiogenic shock | 13 (2.9) | 9 (2.0) | 0.058 | 7 (3.7) | 8 (4.3) | 0.027 |
| Current heart failure | 62 (13.7) | 61 (13.5) | 0.006 | 30 (16.0) | 30 (16.0) | <0.001 |
| Past history and comorbidities | ||||||
| Prior myocardial infarction | 32 (7.1) | 33 (7.3) | 0.009 | 17 (9.0) | 22 (11.7) | 0.087 |
| Prior stroke | 41 (9.1) | 49 (10.9) | 0.059 | 18 (9.6) | 20 (10.6) | 0.035 |
| Prior heart failure | 77 (17.1) | 84 (18.6) | 0.041 | 43 (22.9) | 41 (21.8) | 0.026 |
| Atrial fibrillation | 43 (9.5) | 53 (11.8) | 0.072 | 24 (12.8) | 22 (11.7) | 0.032 |
| Peripheral artery disease | 28 (6.2) | 27 (6.0) | 0.009 | 24 (12.8) | 20 (10.6) | 0.066 |
| Hypertension | 370 (82.0) | 364 (80.7) | 0.034 | 146 (77.7) | 151 (80.3) | 0.065 |
| Hyperlipidemia | 294 (6.5) | 320 (71.0) | 0.124 | 130 (69.1) | 136 (72.3) | 0.070 |
| Diabetes | 210 (46.7) | 201 (44.6) | 0.040 | 91 (48.4) | 84 (44.7) | 0.075 |
| With insulin therapy | 23 (5.1) | 24 (5.3) | 0.010 | 17 (9.0) | 18 (9.6) | 0.018 |
| Current smoker | 99 (22.0) | 97 (21.5) | 0.011 | 26 (13.8) | 33 (17.6) | 0.103 |
| Left ventricular ejection fraction, % | 56.9 ± 10.7 | 57.6 ± 11.1 | 0.062 | 55.8 ± 11.2 | 56.7 ± 12.1 | 0.068 |
| <40 | 22 (4.9) | 23 (5.1) | 0.009 | 15 (8.0) | 14 (7.4) | 0.003 |
| Hemoglobin <11.0 g/dL | 38 (8.4) | 42 (9.3) | 0.031 | 27 (14.4) | 31 (16.5) | 0.059 |
| Platelet count <100∗109/L | 7 (1.6) | 5 (1.1) | 0.039 | 1 (0.5) | 3 (1.6) | 0.104 |
| Moderate chronic kidney disease (eGFR ≤30->60 mL/min/1.73 m2) | 187 (41.5) | 185 (41.0) | 0.009 | 73 (38.8) | 69 (36.7) | 0.044 |
| Severe chronic kidney disease | 40 (8.9) | 43 (9.5) | 0.023 | 40 (21.3) | 39 (20.7) | 0.013 |
| eGFR <30 mL/min/1.73 m2 not on hemodialysis | 17 (3.8) | 20 (4.4) | 0.034 | 13 (6.9) | 13 (6.9) | <0.001 |
| Hemodialysis | 23 (5.1) | 23 (5.1) | <0.001 | 27 (14.4) | 26 (13.8) | 0.015 |
| Cancer history | 50 (11.1) | 49 (10.9) | 0.007 | 20 (10.6) | 26 (13.8) | 0.098 |
| Prior bleeding requring hospitalization or transfusion | 11 (2.4) | 7 (1.6) | 0.063 | 2 (1.1) | 3 (1.6) | 0.046 |
| Liver cirrhosis | 2 (0.4) | 3 (0.7) | 0.030 | 1 (0.5) | 1 (0.5) | <0.001 |
| Long-term use of oral NSAIDs or steroids | 23 (5.1) | 15 (3.3) | 0.088 | 10 (5.3) | 13 (6.9) | 0.067 |
| ARC-HBR | 249 (55.2) | 248 (55.0) | 0.004 | 135 (71.8) | 135 (71.8) | <0.001 |
| Procedural characteristics at the index PCI | ||||||
| Staged PCI | 87 (19.3) | 91 (20.2) | 0.022 | 32 (17.0) | 31 (16.5) | 0.014 |
| Number of target lesions | 1.0 (1.0-1.0) | 1.0 (1.0-1.0) | 0.003 | 1.0 (1.0-2.0) | 1.0 (1.0-2.0) | 0.090 |
| Target of 2 vessels or more | 114 (25.3) | 121 (26.8) | 0.035 | 43 (22.9) | 44 (23.4) | 0.013 |
| Target of chronic total occlusion | 21 (4.7) | 21 (4.7) | <0.001 | 8 (4.3) | 10 (5.3) | 0.050 |
| Target of bifurcation lesion | 147 (32.6) | 141 (31.3) | 0.029 | 50 (26.6) | 52 (27.7) | 0.024 |
| Bifurcation with 2 stents | 7 (1.6) | 7 (1.6) | <0.001 | 1 (0.5) | 2 (1.1) | 0.060 |
| Target of left main coronary artery | 31 (6.9) | 23 (5.1) | 0.075 | 11 (5.9) | 9 (4.8) | 0.047 |
| Target of left anterior descending coronary artery | 280 (62.1) | 272 (60.3) | 0.036 | 101 (53.7) | 108 (57.4) | 0.075 |
| Number of implanted stents | 1.0 (1.0-1.0) | 1.0 (1.0-1.0) | 0.041 | 1.0 (1.0-2.0) | 1.0 (1.0-2.0) | 0.103 |
| Total stent length, mm | 33.0 (23.0-51.0) | 33.0 (23.5-51.0) | 0.041 | 28 (18-48) | 33 (23-53.0) | 0.145 |
| Use of intracoronary imaging | 416 (92.2) | 422 (93.6) | 0.052 | 176 (93.6) | 175 (93.1) | 0.021 |
| Medication at discharge | ||||||
| Antiplatelet agents | 447 (99.1) | 446 (98.9) | 0.022 | 188 (100) | 184 (97.9) | 0.209 |
| Aspirin only | 0 (0.0) | 1 (0.2) | 8 (4.3) | 5 (2.7) | ||
| Prasugrel only | 425 (94.2) | 434 (96.2) | 1 (0.5) | 6 (3.2) | ||
| DAPT | 14 (3.1) | 4 (0.9) | 179 (95.2) | 173 (92.0) | ||
| Anticoagulants | 61 (13.5) | 63 (14.0) | 0.013 | 26 (13.8) | 25 (13.3) | 0.016 |
| Warfarin | 14 (3.1) | 6 (1.3) | 4 (2.1) | 5 (2.7) | ||
| Direct oral anticoagulants | 47 (10.4) | 57 (12.6) | 22 (11.7) | 20 (10.6) | ||
| Renin-angiotensin system inhibitors | 297 (65.9) | 298 (66.1) | 0.005 | 107 (56.9) | 106 (56.4) | 0.011 |
| Mineralocorticoid receptor antagonists | 37 (8.2) | 37 (8.2) | <0.001 | 9 (4.8) | 18 (9.6) | 0.186 |
| Beta-blockers | 227 (50.3) | 240 (53.2) | 0.058 | 88 (46.8) | 91 (48.4) | 0.032 |
| Statins | 403 (89.4) | 403 (89.4) | <0.001 | 164 (87.2) | 161 (85.6) | 0.047 |
| High-intensity statin therapy | 201 (44.6) | 177 (39.2) | 0.108 | 85 (45.2) | 71 (37.8) | 0.152 |
| Histamine H2 receptor blockers | 2 (0.4) | 2 (0.4) | <0.001 | 3 (1.6) | 3 (1.6) | <0.001 |
Values are n (%), mean ± SD, or median (Q1-Q3).
ARC-HBR = Academic Research Consortium for High Bleeding Risk; DAPT = dual antiplatelet therapy; eGFR = estimated glomerular filtration rate; NSAIDs = nonsteroidal anti-inflammatory drugs; PCI = percutaneous coronary intervention; PPI = proton pump inhibitor; SMD = standardized mean difference.
Clinical outcomes
In the propensity score matched cohort with the no-aspirin strategy, the cumulative 1-year incidence of the co-primary cardiovascular endpoint was significantly higher in the PPI group than in the no-PPI group (7.1% vs 2.4%, log-rank P = 0.001), whereas the cumulative 1-year incidence of the co-primary bleeding endpoint was not different between the PPI and no-PPI groups (5.5% vs 3.3%, log-rank P = 0.145) (Table 2, Figure 2A and 2B). The cumulative 1-year incidence of death was higher in the PPI group than in the no-PPI group (4.7% vs 0.9%, log-rank P < 0.001) (Table 2, Figure 2C). The cumulative 1-year incidences of the gastrointestinal bleeding and upper gastrointestinal bleeding were not different between the 2 groups (Table 2).
Table 2.
Clinical Outcomes in the Propensity Matched Cohort
| No-Aspirin Strategy |
Aspirin Strategy |
|||||||
|---|---|---|---|---|---|---|---|---|
| PPI Group | No-PPI Group | HR (95% CI) | P Value | PPI Group | No-PPI Group | HR (95% CI) | P Value | |
| (n = 451) |
(n = 451) |
(n = 188) |
(n = 188) |
|||||
| No. Patients With Event (Cumulative 1-Year Incidence) | No. Patients With Event (Cumulative 1-Year Incidence) | |||||||
| Co-primary cardiovascular endpoint: | ||||||||
| A composite of cardiovascular death, myocardial infarction, definite stent thrombosis, or ischemic stroke | 32 (7.1) | 11 (2.4) | 2.98 (1.50-5.91) | 0.002 | 13 (6.9) | 14 (7.4) | 0.91 (0.43-1.95) | 0.817 |
| Co-primary bleeding endpoint: | ||||||||
| BARC 3 or 5 bleeding | 25 (5.5) | 16 (3.5) | 1.59 (0.85-2.97) | 0.150 | 13 (6.9) | 8 (4.3) | 1.63 (0.67-3.93) | 0.278 |
| Secondary endpoints | ||||||||
| Death | 21 (4.7) | 4 (0.9) | 5.30 (1.82-15.45) | 0.002 | 10 (5.3) | 12 (6.4) | 0.82 (0.35-1.90) | 0.646 |
| Death from cardiovascular causes | 14 (3.1) | 3 (0.7) | 4.71 (1.35-16.38) | 0.015 | 4 (2.1) | 5 (2.7) | 0.74 (0.17-3.32) | 0.698 |
| Death from noncardiovascular causes | 7 (1.6) | 1 (0.2) | 7.07 (0.87-57.47) | 0.067 | 6 (3.2) | 7 (3.7) | 0.84 (0.28-2.51) | 0.758 |
| Myocardial infarction | 14 (3.1) | 6 (1.3) | 2.37 (0.91-6.17) | 0.077 | 3 (1.6) | 5 (2.7) | 0.59 (0.14-2.47) | 0.472 |
| Spontaneous myocardial infarction | 8 (1.8) | 5 (1.1) | 1.63 (0.53-4.99) | 0.389 | 2 (1.1) | 1 (0.5) | 1.93 (0.17-21.3) | 0.592 |
| Procedural myocardial infarction | 6 (1.3) | 1 (0.2) | 6.05 (0.73-50.22) | 0.096 | 1 (0.5) | 4 (2.1) | 0.25 (0.03-2.23) | 0.214 |
| Definite or probable stent thrombosis | 1 (0.2) | 1 (0.2) | 1.0 (0.06-15.97) | 0.999 | 2 (1.1) | 1 (0.5) | 1.99 (0.18-21.9) | 0.575 |
| Definite | 0 (0) | 1 (0.2) | NA | 2 (1.1) | 1 (0.5) | 1.99 (0.18-21.9) | 0.575 | |
| Probable | 1 (0.2) | 0 (0) | NA | 0 (0) | 0 (0) | NA | ||
| Stroke | 8 (1.8) | 3 (0.7) | 2.70 (0.72-10.16) | 0.143 | 5 (2.7) | 5 (2.7) | 1.00 (0.29-3.44) | 1.000 |
| Ischemic stroke | 8 (1.8) | 3 (0.7) | 2.70 (0.72-10.16) | 0.143 | 5 (2.7) | 4 (2.1) | 1.24 (0.33-4.64) | 0.744 |
| Hemorrhagic stroke | 0 (0) | 0 (0) | NA | 0 (0) | 1 (0.5) | NA | ||
| Any coronary revascularization | 31 (6.9) | 12 (2.7) | 2.66 (1.36-5.17) | 0.004 | 14 (7.4) | 6 (3.2) | 2.31 (0.89-6.02) | 0.086 |
| Target-lesion revascularization | 16 (3.5) | 4 (0.9) | 4.09 (1.37-12.25) | 0.012 | 3 (1.6) | 4 (2.1) | 0.74 (0.17-3.31) | 0.693 |
| Non–target-lesion revascularization | 19 (4.2) | 9 (2.0) | 2.16 (0.98-4.77) | 0.057 | 14 (7.4) | 2 (1.1) | 7.04 (1.60-30.96) | 0.010 |
| Bleeding | ||||||||
| BARC 2, 3, or 5 bleeding | 46 (10.2) | 49 (10.9) | 0.93 (0.63-1.40) | 0.760 | 17 (9.0) | 19 (10.1) | 0.89 (0.46-1.71) | 0.719 |
| BARC 5 | 2 (0.4) | 0 (0) | NA | 3 (1.6) | 1 (0.5) | 2.95 (0.31-28.40) | 0.348 | |
| BARC 3 | 23 (5.1) | 16 (3.5) | 1.46 (0.77-2.76) | 0.247 | 10 (5.3) | 7 (3.7) | 1.43 (0.55-3.77) | 0.465 |
| BARC 2 | 23 (5.1) | 36 (8.0) | 0.63 (0.37-1.07) | 0.085 | 6 (3.2) | 12 (6.4) | 0.49 (0.18-1.31) | 0.157 |
| Gastrointestinal bleeding (BARC 3 or 5) | 4 (0.9) | 5 (1.1) | 0.81 (0.22-3.00) | 0.747 | 2 (1.1) | 2 (1.1) | 0.98 (0.14-6.99) | 0.988 |
| Upper | 0 (0) | 1 (0.2) | NA | 0 (0) | 1 (0.5) | NA | ||
| Lower | 3 (0.7) | 2 (0.4) | 1.51 (0.25-9.04) | 0.651 | 1 (0.5) | 1 (0.5) | 0.99 (0.06-15.77) | 0.992 |
| Unknown | 1 (0.2) | 2 (0.4) | 0.58 (0.05-5.56) | 0.577 | 1 (0.5) | 0 (0) | NA | |
| Gastrointestinal bleeding (BARC 2, 3, or 5) | 10 (2.2) | 8 (1.8) | 1.27 (0.50-3.21) | 0.619 | 4 (2.1) | 3 (1.6) | 1.32 (0.30-5.89) | 0.717 |
| Upper | 2 (0.4) | 1 (0.2) | 2.04 (0.18-22.45) | 0.562 | 1 (0.5) | 1 (0.5) | 0.99 (0.06-15.90) | 0.997 |
| Lower | 6 (1.3) | 5 (1.1) | 1.21 (0.37-3.98) | 0.748 | 1 (0.5) | 2 (1.1) | 0.49 (0.04-5.45) | 0.565 |
| Unknown | 2 (0.4) | 2 (0.4) | 1.01 (0.14-7.17) | 0.992 | 2 (1.1) | 0 (0) | NA | |
NA = not available; other abbreviations as in Table 1.
Figure 2.
Cumulative Incidence in Patients With the No-Aspirin Strategy
The cumulative incidence was estimated with the Kaplan-Meier method, and the difference was compared with the log-rank test in the patients with the no-aspirin strategy. (A) Co-primary cardiovascular endpoint: a composite of cardiovascular death, myocardial infarction, definite stent thrombosis, or ischemic stroke in no-aspirin group. (B) Co-primary bleeding endpoint: BARC 3 or 5 bleeding in the no-aspirin group. (C) Death. BARC = Bleeding Academic Research Consortium; other abbreviations as in Figure 1.
In the propensity score matched cohort with the aspirin strategy, the cumulative 1-year incidences of the co-primary cardiovascular and bleeding endpoints were not different between the PPI and no-PPI groups (6.9% vs 7.4%, log-rank P = 0.817, and 6.9% vs 4.3%, log-rank P = 0.278) (Table 2, Figure 3A and 3B). The cumulative 1-year incidence of death was also not different between the 2 groups (5.3% vs 6.4%, log-rank P = 0.646) (Figure 3C). The cumulative 1-year incidence of coronary revascularization was higher in the PPI group than in the no-PPI group (Table 2). The cumulative 1-year incidences of gastrointestinal bleeding and upper gastrointestinal bleeding were not different between the 2 groups (Table 2).
Figure 3.
Cumulative Incidence in Patients With the Aspirin Strategy
The cumulative incidence was estimated with the Kaplan-Meier method, and the difference was compared with the log-rank test in patients with the aspirin strategy. (A) Co-primary cardiovascular endpoint: a composite of cardiovascular death, myocardial infarction, definite stent thrombosis, or ischemic stroke in the aspirin group. (B) Co-primary bleeding endpoint: BARC 3 or 5 bleeding in the aspirin group. (C) Death. Abbreviations as in Figures 1 and 2.
In the sensitivity analyses using the multivariable Cox proportional hazard models in the entire cohort, the results were fully consistent with those in the propensity score matching analyses (Supplemental Tables 2 and 3). In the sensitivity analyses in which the time of the discharge was regarded as time 0, the results were fully consistent with those in the main analyses (Supplemental Table 4).
Discussion
The main findings in the present study were as follows: 1) PPI use was not associated with a lower risk of major bleeding regardless of the no-aspirin and aspirin strategies; 2) PPI use was associated with a higher risk of cardiovascular events and mortality with the no-aspirin strategy, but not with the aspirin strategy (Central Illustration).
Central Illustration.
Effect of Proton Pump Inhibitor in the No-Aspirin Strategy and the Aspirin Strategy
In the propensity score matched cohort, cardiovascular event and death more often occurred in the PPI group than in the no-PPI group with the no-aspirin strategy, but not with the aspirin strategy. Incidence of major bleeding was not different between the groups regardless of the no-aspirin and aspirin strategies. PPI use increased cardiovascular events and mortality without reducing major bleeding in aspirin-free P2Y12 monotherapy. 1M = 1 month; CV event = cardiovascular event; DAPT = dual antiplatelet therapy; PCI = percutaneous coronary intervention; PPI = proton pump inhibitor; STOPDAPT-3 = ShorT and OPtimal duration of Dual AntiPlatelet Therapy after everolimus-eluting cobalt-chromium stent-3.
COGENT (Clopidogrel and the Optimization of Gastrointestinal Events Trial) demonstrated that PPIs reduced upper gastrointestinal bleeding in patients with coronary artery disease receiving DAPT with clopidogrel and aspirin.4 The American Heart Association and European Society of Cardiology guidelines stated that PPIs were recommended to reduce upper gastrointestinal bleeding in patients receiving DAPT.17,18 On the other hand, the European Society of Cardiology guideline stated that PPIs in patients with single antiplatelet therapy should be used under consideration of the individual gastrointestinal bleeding risk, because there was a scarcity of data on the efficacy of PPIs in reducing gastrointestinal bleeding in patients receiving single antiplatelet therapy, especially P2Y12 inhibitor monotherapy.18 In the present study, PPI use was not associated with a lower incidence of major bleeding, including upper gastrointestinal bleeding, with the no-aspirin strategy. An observational study from the GLOBAL LEADERS (A Clinical Study Comparing Two Forms of Anti-platelet Therapy After Stent Implantation) trial reported that PPI use was not associated with a lower incidence of major bleeding in patients receiving ticagrelor monotherapy (gastrointestinal bleeding not evaluated), which was consistent with the present study.19 When patients took only P2Y12 inhibitor monotherapy, not DAPT, benefit of PPIs in reducing gastrointestinal bleeding might be attenuated. However, several limitations might be considered for the lack of difference in bleeding between patients with and without PPI prescription in the present study. First, the present observational study could not exclude the effect of unmeasured confounders. Second, the present study might not have adequate power to detect the benefit of PPIs due to very low incidence of upper gastrointestinal bleeding and short duration of follow-up (only 1 year). Indeed, the benefit of PPIs in reducing upper gastrointestinal bleeding was not observed with the aspirin strategy in the present study, which was discordant with the results in the previous randomized clinical trials demonstrating the benefit of PPIs in reducing upper gastrointestinal bleeding in patients receiving aspirin monotherapy.3,5 Helicobacter pylori infection was reported to increase the risk of gastrointestinal bleeding in patients receiving aspirin or non-aspirin antiplatelet agents.20 Recently, H pylori infection has decreased in real clinical practice, which might be one of the reasons for the very low incidence of gastrointestinal bleeding in the present study.21 Clopidogrel monotherapy was known to have a lower risk of gastrointestinal bleeding compared with aspirin monotherapy.22 However, the randomized OPT-PEACE (Optimal Antiplatelet Therapy for Prevention of Gastrointestinal Injury Evaluated by Ankon Magnetically Controlled Capsule Endoscopy) trial showed that clopidogrel monotherapy had comparable risk of gastrointestinal mucosal injury detected on capsule endoscopy compared with aspirin monotherapy.23 In one randomized clinical trial that enrolled patients with a history of aspirin-induced upper gastrointestinal bleeding, clopidogrel with PPIs placebo compared with aspirin with PPIs had higher incidence of recurrent upper gastrointestinal bleeding.24 Further studies would be warranted to evauate the need of prophylactic use of PPIs in patients receiving P2Y12 inhibitor monotherapy.
Previous studies showed that PPIs might reduce an antiplatelet effect of clopidogrel, which was mediated by P2Y12 receptors, primarily by inhibiting hepatic cytochrome P450 2C19 (CYP2C19).9,10 The attenuation of antiplatelet effect by PPIs was also oberseved in patients receiving prasugrel and ticagrelor.11,12 There were conflicting data regarding the risk of cardiovascular events in patients taking both P2Y12 inhibitors and PPIs.4,6,9, 10, 11, 12,19 However, these data were derived from patients with concomitant use of aspirin, and there was a scarcity of data on the safety of PPIs in patients receiving P2Y12 inhibitor monotherapy. An observational study from the GLOBAL LEADERS trial reported that PPI use was not associated with higher incidence of cardiovascular events in patients receiving ticagrelor monotherapy.19 On the other hand, PPI use was asscociated with a higher incidence of cardiovascular events with the no-aspirin strategy, but not with the aspirin strategy in the present study. Because of the nature of the observational study design, short duration of follow-up (only 1 year), and the discordant results in the preceding 2 studies, there remains uncertainty regarding the association of PPIs with an increase in cardiovascular events in patients receiving P2Y12 inhibitor monotherapy. However, the association between PPI use and increased risk of cardiovascular events with the no-aspirin strategy in the present study might raise concern on the safety of PPIs in patients receiving P2Y12 inhibitor monotherapy. Several observational studies suggested that PPI use might increase not only cardiovascular events, but also noncardiovascular events such as pneumonia, fracture, Clostridium difficile infection, enteric infection, or dementia.25, 26, 27, 28 However, in the COMPASS (Cardiovascular Outcomes for People Using Anticoagulation Strategy) trial, which enrolled patients receiving aspirin or rivaroxaban, PPI use did not increase noncardiovascular adverse events including death, noncardiovascular death, pneumonia, fracture, C difficile infection, or dementia, whereas enteric infection was higher in patients with PPIs than in those with placebo during 3-year follow-up.29 In the present study, noncardiovascular death was numerically higher in patients with PPIs than in those without PPIs in the no-aspirin strategy, but not in the aspirin strategy, which could be mechanistically difficult to interpret. Further studies would be warranted to explore the cardiovascular and noncardiovascular safety of PPIs in patients receiving P2Y12 inhibitor or aspirin monotherapy.
Study limitations
The present study had several important limitations. First, the present study was not a randomized controlled trial for PPI prescription. Despite propensity score matching and multivariable analyses, selection bias and residual confounding would be inevitable. Second, the present study was underpowered to evaluate the effect of PPI prescription due to the small number of patients with event and short duration of follow-up. Indeed, the higher rate of the co-primary cardiovascular event in the PPI group than in the no-PPI group with the no-aspirin strategy was driven by the very low event rate in the no-PPI group. The event rate in the PPI group with the no-aspirin strategy was apparently similar to the event rates in the PPI and no-PPI groups with the aspirin strategy. Therefore, the higher event rate in the PPI group than in the no-PPI group with the no-aspirin strategy could be a chance finding in this underpowered comparison. In addition, the benefit of PPIs in reducing upper gastrointestinal bleeding was not observed with the aspirin strategy in the present study, although the previous randomized clinical trials demonstrated the benefit of PPIs in reducing upper gastrointestinal bleeding in patients receiving aspirin monotherapy.3,5 This observation might also suggest that the present study did not have sufficient statistical power to identify the true effect of PPIs. Third, PPI use was evaluated only at the time of discharge from the index hospitalization, not at the time of randomization. Therefore, we excluded patients who died during index hospitalization in the present study population. The effect of immortal time bias associated with the excluding of patients with death during index hospitalization should be considered. However, the underlying immortal time bias might not be so large, because the results in the sensitivity analyses in which the time of the discharge was regarded as time 0 were fully consistent with those in the main analyses. Fourth, we did not collect data on discontinuation or noncompliance of PPIs.
Conclusions
PPI use was associated with a higher risk of cardiovascular events and mortality without decreasing major bleeding in patients undergoing PCI with an aspirin-free P2Y12 inhibitor monotherapy.
Funding Support and Author Disclosures
The source of funding was Abbott Medical Japan. Dr Yamamoto has received honoraria from Abbott Medical Japan, Daiichi-Sankyo, and Boston Scientific. Dr Natsuaki has received honoraria from Abbott Medical Japan, Daiichi-Sankyo, Medtronic, Terumo, Japan Lifeline, Asahi Intecc, Bristol-Myers Squibb, Otsuka, Amgen, Sanofi, Takeda and Bayer. Dr Watanabe has received personal fees from Abbott Medical Japan during the conduct of the study as well as personal fees from Daiichi-Sankyo, Kowa, Abiomed, Bayer, Pfizer, Bristol-Myers Squibb, and Otsuka outside the submitted work. Dr Morimoto has received lecturer's fees from Abbott, AstraZeneca, Boehringer Ingelheim, Bristol-Myers Squibb, Daiichi-Sankyo, Japan Lifeline, Pfizer, Tsumura, and UCB; manuscript fees from Pfizer; and served on the advisory board for GlaxoSmithKline, Novartis, and Teijin. Dr Suwa has received personal fees from Abbott Medical Japan and Daiichi-Sankyo outside the submitted work. Dr Kimura has received grants from Abbott Medical Japan and Boston Scientific and has been an advisory board member of Abbott Medical Japan and Terumo Japan. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
Acknowledgments
The authors appreciate the study investigators for their efforts in enrolling patients and collecting data. They also appreciate the members of the Research Institute for Production Development for coordinating the study.
Footnotes
The authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the Author Center.
Appendix
For study organization, participating centers in the study, endpoints evaluated in 1-year analysis, definition of endpoints and, supplemental methods and tables, please see the online version of this paper.
Appendix
References
- 1.Généreux P., Giustino G., Witzenbichler B., et al. Incidence, predictors, and impact of post-discharge bleeding after percutaneous coronary intervention. J Am Coll Cardiol. 2015;66:1036–1045. doi: 10.1016/j.jacc.2015.06.1323. [DOI] [PubMed] [Google Scholar]
- 2.Yamamoto K., Natsuaki M., Morimoto T., et al. Ischemic and bleeding events after first major bleeding event in patients undergoing coronary stent implantation. Am J Cardiol. 2022;162:13–23. doi: 10.1016/j.amjcard.2021.09.017. [DOI] [PubMed] [Google Scholar]
- 3.Lai K.C., Lam S.K., Chu K.M., et al. Lansoprazole for the prevention of recurrences of ulcer complications from long-term low-dose aspirin use. N Engl J Med. 2002;346:2033–2038. doi: 10.1056/NEJMoa012877. [DOI] [PubMed] [Google Scholar]
- 4.Bhatt D.L., Cryer B.L., Contant C.F., et al. Clopidogrel with or without omeprazole in coronary artery disease. N Engl J Med. 2010;363:1909–1917. doi: 10.1056/NEJMoa1007964. [DOI] [PubMed] [Google Scholar]
- 5.Moayyedi P., Eikelboom J.W., Bosch J., et al. Pantoprazole to prevent gastroduodenal events in patients receiving rivaroxaban and/or aspirin in a randomized, double-blind, placebo-controlled trial. Gastroenterology. 2019;157:403–412.e5. doi: 10.1053/j.gastro.2019.04.041. [DOI] [PubMed] [Google Scholar]
- 6.Yamamoto K., Shiomi H., Morimoto T., et al. Comparison of the OPTIVUS-Complex PCI multivessel cohort with the historical CREDO-Kyoto Registry Cohort-3. Circ J. 2023;87:1689–1702. doi: 10.1253/circj.CJ-22-0837. [DOI] [PubMed] [Google Scholar]
- 7.Watanabe H., Domei T., Morimoto T., et al. Effect of 1-month dual antiplatelet therapy followed by clopidogrel vs 12-month dual antiplatelet therapy on cardiovascular and bleeding events in patients receiving PCI: the STOPDAPT-2 randomized clinical trial. JAMA. 2019;321:2414–2427. doi: 10.1001/jama.2019.8145. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Mehran R., Baber U., Sharma S.K., et al. Ticagrelor with or without aspirin in high-risk patients after PCI. N Engl J Med. 2019;381:2032–2042. doi: 10.1056/NEJMoa1908419. [DOI] [PubMed] [Google Scholar]
- 9.Weisz G., Smilowitz N.R., Kirtane A.J., et al. Proton pump inhibitors, platelet reactivity, and cardiovascular outcomes after drug-eluting stents in clopidogrel-treated patients: the ADAPT-DES study. Circ Cardiovasc Interv. 2015;8 doi: 10.1161/CIRCINTERVENTIONS.114.001952. [DOI] [PubMed] [Google Scholar]
- 10.Gilard M., Arnaud B., Cornily J.C., 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;51:256–260. doi: 10.1016/j.jacc.2007.06.064. [DOI] [PubMed] [Google Scholar]
- 11.Goodman S.G., Clare R., Pieper K.S., et al. Association of proton pump inhibitor use on cardiovascular outcomes with clopidogrel and ticagrelor: insights from the platelet inhibition and patient outcomes trial. Circulation. 2012;125:978–986. doi: 10.1161/CIRCULATIONAHA.111.032912. [DOI] [PubMed] [Google Scholar]
- 12.O'Donoghue M.L., Braunwald E., Antman E.M., 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;374:989–997. doi: 10.1016/S0140-6736(09)61525-7. [DOI] [PubMed] [Google Scholar]
- 13.Natsuaki M., Watanabe H., Morimoto T., et al. An aspirin-free versus dual antiplatelet strategy for coronary stenting: STOPDAPT-3 randomized trial. Circulation. 2024;149(8):585–600. doi: 10.1161/CIRCULATIONAHA.123.066720. [DOI] [PubMed] [Google Scholar]
- 14.Watanabe H., Natsuaki M., Morimoto T., et al. Aspirin versus clopidogrel monotherapy after percutaneous coronary intervention: 1-year follow-up of the STOPDAPT-3 trial. Eur Heart J. 2025;18(17):2120–2135. doi: 10.1093/eurheartj/ehae617. [DOI] [PubMed] [Google Scholar]
- 15.Mehran R., Rao S.V., Bhatt D.L., et al. Standardized bleeding definitions for cardiovascular clinical trials: a consensus report from the Bleeding Academic Research Consortium. Circulation. 2011;123:2736–2747. doi: 10.1161/CIRCULATIONAHA.110.009449. [DOI] [PubMed] [Google Scholar]
- 16.Cutlip D.E., Windecker S., Mehran R., et al. Clinical end points in coronary stent trials: a case for standardized definitions. Circulation. 2007;115:2344–2351. doi: 10.1161/CIRCULATIONAHA.106.685313. [DOI] [PubMed] [Google Scholar]
- 17.Virani S.S., Newby L.K., Arnold S.V., et al. 2023 AHA/ACC/ACCP/ASPC/NLA/PCNA guideline for the management of patients with chronic coronary disease: a report of the American Heart Association/American College of Cardiology Joint Committee on Clinical Practice Guidelines. J Am Coll Cardiol. 2023;82(9):833–955. doi: 10.1016/j.jacc.2023.04.003. [DOI] [PubMed] [Google Scholar]
- 18.Vrints C., Andreotti F., Koskinas K.C., et al. 2024 ESC guidelines for the management of chronic coronary syndromes. Eur Heart J. 2024;45:3415–3537. doi: 10.1093/eurheartj/ehae177. [DOI] [PubMed] [Google Scholar]
- 19.Ono M., Onuma Y., Kawashima H., et al. Impact of proton pump inhibitors on efficacy of antiplatelet strategies with ticagrelor or aspirin after percutaneous coronary intervention: insights from the GLOBAL LEADERS trial. Catheter Cardiovasc Interv. 2022;100:72–82. doi: 10.1002/ccd.30217. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Venerito M., Schneider C., Costanzo R., et al. Contribution of Helicobacter pylori infection to the risk of peptic ulcer bleeding in patients on nonsteroidal anti-inflammatory drugs, antiplatelet agents, anticoagulants, corticosteroids and selective serotonin reuptake inhibitors. Aliment Pharmacol Ther. 2018;47:1464–1471. doi: 10.1111/apt.14652. [DOI] [PubMed] [Google Scholar]
- 21.Li Y., Choi H., Leung K., et al. Global prevalence of Helicobacter pylori infection between 1980 and 2022: a systematic review and meta-analysis. Lancet Gastroenterol Hepatol. 2023;8:553–564. doi: 10.1016/S2468-1253(23)00070-5. [DOI] [PubMed] [Google Scholar]
- 22.Gragnano F., Cao D., Pirondini L., et al. P2Y12 inhibitor or aspirin monotherapy for secondary prevention of coronary events. J Am Coll Cardiol. 2023;82:89–105. doi: 10.1016/j.jacc.2023.04.051. [DOI] [PubMed] [Google Scholar]
- 23.Han Y., Liao Z., Li Y., et al. Magnetically controlled capsule endoscopy for assessment of antiplatelet therapy-induced gastrointestinal injury. J Am Coll Cardiol. 2022;79:116–128. doi: 10.1016/j.jacc.2021.10.028. [DOI] [PubMed] [Google Scholar]
- 24.Chan F.K., Ching J.Y., Hung L.C., et al. Clopidogrel versus aspirin and esomeprazole to prevent recurrent ulcer bleeding. N Engl J Med. 2005;352:238–244. doi: 10.1056/NEJMoa042087. [DOI] [PubMed] [Google Scholar]
- 25.Laheij R.J., Sturkenboom M.C., Hassing R.J., et al. Risk of community-acquired pneumonia and use of gastric acid-suppressive drugs. JAMA. 2004;292:1955–1960. doi: 10.1001/jama.292.16.1955. [DOI] [PubMed] [Google Scholar]
- 26.Yang Y.X., Lewis J.D., Epstein S., et al. Long-term proton pump inhibitor therapy and risk of hip fracture. JAMA. 2006;296:2947–2953. doi: 10.1001/jama.296.24.2947. [DOI] [PubMed] [Google Scholar]
- 27.Dial S., Delaney J.A., Barkun A.N., et al. Use of gastric acid-suppressive agents and the risk of community-acquired Clostridium difficile-associated disease. JAMA. 2005;294:2989–2995. doi: 10.1001/jama.294.23.2989. [DOI] [PubMed] [Google Scholar]
- 28.Gomm W., von Holt K., Thomé F., et al. Association of proton pump inhibitors with risk of dementia: a pharmacoepidemiological claims data analysis. JAMA Neurol. 2016;73:410–416. doi: 10.1001/jamaneurol.2015.4791. [DOI] [PubMed] [Google Scholar]
- 29.Moayyedi P., Eikelboom J.W., Bosch J., et al. Safety of proton pump inhibitors based on a large, multi-year, randomized trial of patients receiving rivaroxaban or aspirin. Gastroenterology. 2019;157:682–691.e2. doi: 10.1053/j.gastro.2019.05.056. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.