Abstract
There is little evidence on whether non‐steroidal anti‐inflammatory drugs (NSAIDs) and aspirin interact in secondary cardiovascular prevention in type 2 diabetic patients. This is an observational study using data from the Action to Control Cardiovascular Risk in Diabetes and Follow‐on studies. Hazard ratios (HRs) for mortality with 95% confidence intervals (95% CIs) were calculated using Cox proportional hazard models to compare time to death in patients using and not using aspirin who were simultaneously using or not using NSAIDs. A total of 3600 type 2 diabetic patients with cardiovascular disease were included. During a mean follow‐up period of 8.8 years, 948 patients died. After adjustments, the risk of all‐cause mortality in patients not using NSAIDs was significantly lower in those using aspirin than in those not using aspirin (HR, 0.81; 95% CI, 0.70‐0.93; P = 0.004). The risk in patients using NSAIDs did not differ significantly between the two groups. There was a significant interaction between aspirin use and NSAIDs use. In type 2 diabetic patients with cardiovascular disease, aspirin use was not beneficial for those using NSAIDs.
Keywords: ACCORD, anti‐inflammatory drugs, aspirin, cardiovascular events, mortality, non‐cardiovascular mortality, non‐steroidal secondary prevention
1. INTRODUCTION
The benefits of aspirin use in patients with a history of cardiovascular disease are well‐established in the secondary prevention of cardiovascular events.1 Non‐steroidal anti‐inflammatory drugs (NSAIDs), including nonselective and cyclooxygenase (COX)‐2‐selective NSAIDs, increase the risk of cardiovascular and gastrointestinal events.2, 3, 4 In addition, a previous study had suggested that both nonselective and COX‐2‐selective NSAIDs increase the risk of death in patients with a history of cardiovascular disease.5 Moreover, there are pharmacodynamic concerns regarding concomitant use of NSAIDs in patients using aspirin because nonselective and COX‐2‐selective NSAIDs, such as ibuprofen and celecoxib, may block the access of aspirin to the acetylation site of platelet‐expressed COX‐1.6, 7 However, the clinical data regarding this interaction are inconsistent. Therefore, the aim of this study was to assess whether the concomitant use of aspirin and other NSAIDs may adversely affect the risk of mortality in diabetic patients with a history of cardiovascular disease. Because aspirin and NSAIDs may be associated with non‐cardiovascular events such as cancer and gastrointestinal events, for example, gastrointestinal bleeding and renal failure, we further evaluated the associations between concomitant use of aspirin and NSAIDs and both cardiovascular and non‐cardiovascular mortality.
2. MATERIALS AND METHODS
2.1. Study design and patients
We used data from the Action to Control Cardiovascular Risk in Diabetes (ACCORD)8 and the Action to Control Cardiovascular Risk in Diabetes Follow‐on (ACCORDION)9 studies to assess whether aspirin use is equally effective in type 2 diabetic patients with a history of cardiovascular disease who are using and not using NSAIDs. NSAIDs in the present study did not include aspirin. The study design and patient characteristics of the ACCORD and ACCORDION studies have been previously reported. In the present study, we limited inclusion to patients with a history of clinical cardiovascular disease (n = 3611), which was defined as myocardial infarction, angina pectoris, coronary revascularization including coronary artery bypass grafting and percutaneous coronary intervention, stroke or other revascularization, such as carotid artery revascularization and peripheral artery revascularization. Patients with missing information regarding the use of aspirin and NSAIDs were excluded (n = 11), which resulted in a final sample of 3600 patients. The institutional review board of the National Center for Global Health and Medicine approved the present study. The National Heart, Lung, and Blood Institute (NHLBI) approved the use of ACCORD data.
2.2. Outcomes and measurements
The primary outcome was all‐cause mortality. Secondary outcomes were cardiovascular and non‐cardiovascular mortality and major adverse cardiovascular events (MACE). Cardiovascular death was defined as presumed cardiovascular death, unexpected death and death from myocardial infarction, congestive heart failure, arrhythmia, stroke and other cardiovascular diseases, including pulmonary emboli and abdominal aortic aneurysm rupture. MACE was defined as cardiovascular death, nonfatal myocardial infarction and nonfatal stroke. Participants were followed for a maximum of 13 years.
All prescribed medications that participants were using regularly, including aspirin and NSAIDs, were confirmed at baseline. NSAIDs comprised both nonselective and COX‐2‐selective NSAIDs.
2.3. Statistical analysis
Comparisons were made between patients using and those not using aspirin. Hazard ratios (HRs) for mortality, with 95% confidence intervals (CIs), were calculated using Cox proportional hazard models to compare the time to occurrence of outcome events in the subgroups of aspirin users and nonusers, separately in patients using and not using NSAIDs. To confirm the robustness of results, several multivariable adjustments were made. In addition, propensity score‐matched analyses were performed to confirm results of the primary outcome. All statistical analyses were conducted using Stata software (version 14.1, Stata Corp, College Station, Texas). A P value of <0.05 was considered statistically significant for all tests.
3. RESULTS
3.1. Baseline characteristics
The present study included 3197 patients who were not using NSAIDs and 403 patients who were using NSAIDs. Among patients using NSAIDs, those using aspirin were associated with a lower proportion of women, more use of statins, and lower levels of glycated haemoglobin and low‐density lipoprotein cholesterol; other characteristics were similar between those using and not using aspirin (Table 1).
Table 1.
Baseline characteristics of type 2 diabetic patients with a history of clinical cardiovascular disease using and not using nonsteroidal anti‐inflammatory drugs
| NSAIDs (−) | NSAIDs (+) | |||||
|---|---|---|---|---|---|---|
| Characteristics | Aspirin (−) | Aspirin (+) | P value | Aspirin (−) | Aspirin (+) | P value |
| Number | 1044 | 2153 | 136 | 267 | ||
| Age (years) | 62.3 (8.3) | 62.8 (7.8) | 0.080 | 63.1 (7.7) | 62.5 (6.6) | 0.413 |
| Female sex (%) | 30.5 | 26.0 | 0.008 | 41.2 | 31.1 | 0.044 |
| Race/ethnicity (%) | 0.001 | 0.250 | ||||
| White | 61.1 | 67.8 | 66.2 | 67.4 | ||
| Black | 17.7 | 14.4 | 24.3 | 18.0 | ||
| Hispanic | 9.0 | 6.4 | 5.9 | 7.1 | ||
| Others | 12.2 | 11.3 | 3.7 | 7.5 | ||
| Educational level (%) | 0.059 | 0.452 | ||||
| Less than high school | 17.3 | 14.8 | 14.1 | 18.0 | ||
| High school | 24.2 | 27.4 | 31.9 | 25.5 | ||
| Some college | 34.8 | 32.6 | 37.0 | 36.3 | ||
| College degree or higher | 23.7 | 25.2 | 17.0 | 20.2 | ||
| Currently smoking (%) | 13.6 | 11.9 | 0.169 | 22.8 | 15.0 | 0.052 |
| Body mass index (kg/m2)a | 31.7 (5.3) | 32.0 (5.2) | 0.085 | 33.6 (5.3) | 33.3 (5.1) | 0.572 |
| Duration of diabetes (years) | 11.3 (8.3) | 11.7 (8.0) | 0.208 | 12.2 (8.0) | 12.6 (8.1) | 0.699 |
| Hypertension (%) | 96.7 | 96.3 | 0.605 | 97.1 | 94.8 | 0.290 |
| Dyslipidaemia (%) | 96.9 | 98.2 | 0.024 | 98.5 | 99.3 | 0.490 |
| Medications (%) | ||||||
| Insulin | 39.4 | 40.2 | 0.684 | 50.7 | 43.2 | 0.153 |
| Metformin | 57.9 | 63.5 | 0.002 | 57.4 | 61.1 | 0.474 |
| Thiazolidinedione | 22.6 | 20.6 | 0.199 | 18.4 | 20.2 | 0.660 |
| ACE‐I/ARB | 71.2 | 73.3 | 0.207 | 76.5 | 72.3 | 0.367 |
| CCB | 11.1 | 14.5 | 0.008 | 16.2 | 12.7 | 0.34 |
| Thiazide | 23.3 | 25.6 | 0.155 | 31.6 | 26.6 | 0.289 |
| Beta‐blocker | 43.5 | 57.5 | <0.001 | 49.3 | 59.2 | 0.058 |
| Statin | 71.5 | 79.9 | <0.001 | 63.2 | 79.8 | <0.001 |
| Antiplatelet agents (except aspirin) | 13.8 | 10.4 | 0.004 | 13.2 | 13.5 | 0.945 |
| Glycated haemoglobin (%) | 8.3 (1.1) | 8.3 (1.0) | 0.632 | 8.5 (1.1) | 8.2 (1.0) | 0.019 |
| LDL cholesterol (mg/dL) | 103.2 (33.4) | 97.5 (31.5) | <0.001 | 105.8 (30.5) | 97.7 (29.6) | 0.010 |
| HDL cholesterol (mg/dL) | 40.1 (10.5) | 39.3 (9.9) | 0.043 | 40.8 (11.5) | 39.2 (10.7) | 0.155 |
| Triglyceride (mg/dL) | 192.8 (127.0) | 189.1 (119.0) | 0.422 | 199.7 (126.9) | 190.3 (108.5) | 0.441 |
| eGFR (mL/min/1.73 m2)b | 88.5 (23.4) | 88.3 (22.6) | 0.813 | 88.5 (22.6) | 89.3 (22.8) | 0.712 |
| Systolic blood pressure (mm Hg) | 137.7 (16.5) | 134.4 (17.3) | <0.001 | 136.9 (15.9) | 135.6 (17.1) | 0.446 |
| Diastolic blood pressure (mm Hg) | 74.0 (10.7) | 72.2 (10.6) | <0.001 | 75.0 (10.4) | 73.3 (10.0) | 0.103 |
| Glycaemic control strategy | ||||||
| Intensive glycaemic treatment (%) | 50.7 | 50.4 | 0.884 | 47.8 | 52.4 | 0.378 |
| BP control strategy | ||||||
| Intensive BP treatment (%) | 23.6 | 21.7 | 0.233 | 23.5 | 21.7 | 0.681 |
| Lipid control strategy | ||||||
| Intensive lipid treatment (%) | 25.4 | 29.3 | 0.022 | 22.1 | 29.6 | 0.108 |
Data are presented as number of participants, percent or mean (standard deviation). P value was calculated by comparing variables in aspirin users with those in aspirin nonusers. Categorical variables were compared using chi‐squared tests, and continuous variables were compared using t‐tests.
HbA1c: 8.3% = 67 mmoL/moL; 8.5% = 69 mmoL/moL; 8.2% = 66 mmoL/moL.
Abbreviations: ACE‐I, angiotensin‐converting enzyme inhibitors; ARB, angiotensin II receptor blockers; BP, blood pressure; CCB, calcium channel blockers; eGFR, estimated glomerular filtration rate; HDL, high‐density lipoprotein; LDL, low‐density lipoprotein; NSAIDs, non‐steroidal anti‐inflammatory drugs.
Body mass index was calculated as weight in kilograms divided by square of height in meters.
Estimated GFR was calculated using the following Modification of Diet in Renal Disease Study equation: estimated GFR (mL/min/1.73 m2) = 175 × (serum creatinine in mg/dL)−1.154 × (age in years)−0.203 × (0.742 for female) × (1.212 for African American).
3.2. All‐cause, cardiovascular and non‐cardiovascular mortality
The overall mean (SD) follow‐up period was 8.8 (2.5) years and 944 patients died. All‐cause mortality rates (per 1000 person‐year) were 30.0 and 28.0 in patients not using and those using NSAIDS, respectively. Among patients not using NSAIDs, the risk of all‐cause mortality was significantly lower in those using aspirin than in those not using aspirin (Figure S1). Among patients using NSAIDs, the risk of all‐cause mortality did not differ significantly between those using and those not using aspirin. After multivariable adjustments, the risk of all‐cause mortality in patients not using NSAIDs was significantly lower in those using aspirin than in those not using aspirin (adjusted HR in model 1, 0.81; 95% CI, 0.70‐0.93; P = 0.003; adjusted HR in model 2, 0.80; 95% CI, 0.69‐0.93; P = 0.002; adjusted HR in model 3, 0.80; 95% CI, 0.69‐0.93; P = 0.003) (Table 2). In patients using NSAIDs, the adjusted risk of all‐cause mortality did not differ significantly between those using and those not using aspirin. There were significant interactions between the use of aspirin and the use of NSAIDs in all three multivariable models (P for all interactions <0.05). Propensity score‐matched analyses with well‐balanced baseline characteristics between the two groups did not alter the results (Table S1 and Figure S2). In addition, the risk of all‐cause mortality in patients not using nonselective NSAIDs was significantly lower in those using aspirin than in those not using aspirin (adjusted HR in model 3, 0.81; 95% CI, 0.70–0.94; P = 0.004), whereas that in patients using nonselective NSAIDs did not differ significantly between those using and those not using aspirin (adjusted HR in model 3, 1.51; 95% CI, 0.84‐2.72; P = 0.16) (P for interaction <0.05).
Table 2.
All‐cause, cardiovascular and non‐cardiovascular mortality in type 2 diabetic patients with a history of clinical cardiovascular disease using and not using nonsteroidal anti‐inflammatory drugs
| NSAIDs (−) | NSAIDs (+) | |||||
|---|---|---|---|---|---|---|
| Event | Aspirin (−) | Aspirin (+) | P value | Aspirin (−) | Aspirin (+) | P value |
| In patients with history of clinical CVD a | (n = 1044) | (n = 2153) | (n = 136) | (n = 267) | ||
| All‐cause death | ||||||
| Number of events | 298 | 544 | 32 | 70 | ||
| Event rate (per 1000 person‐year) | 33.3 | 28.5 | 25.8 | 29.1 | ||
| Unadjusted hazard ratio (95% CI) | 1.00 (ref) | 0.84 (0.73–0.97) | 0.01 | 1.00 (ref) | 1.14 (0.75–1.73) | 0.54 |
| Adjusted hazard ratio (95% CI), model 1 | 1.00 (ref) | 0.81 (0.70–0.93) | 0.003 | 1.00 (ref) | 1.28 (0.83–1.98) | 0.25 |
| Adjusted hazard ratio (95% CI), model 2 | 1.00 (ref) | 0.80 (0.69–0.93) | 0.002 | 1.00 (ref) | 1.40 (0.90–2.18) | 0.13 |
| Adjusted hazard ratio (95% CI), model 3 | 1.00 (ref) | 0.80 (0.69–0.93) | 0.003 | 1.00 (ref) | 1.35 (0.85–2.13) | 0.20 |
| Cardiovascular death | ||||||
| Number of events | 125 | 224 | 14 | 29 | ||
| Event rate (per 1000 person‐year) | 14.0 | 11.7 | 11.3 | 12.1 | ||
| Unadjusted hazard ratio (95% CI) | 1.00 (ref) | 0.83 (0.67–1.04) | 0.10 | 1.00 (ref) | 1.07 (0.57–2.03) | 0.82 |
| Adjusted hazard ratio (95% CI), model 1 | 1.00 (ref) | 0.78 (0.63–0.98) | 0.02 | 1.00 (ref) | 1.16 (0.60–2.23) | 0.65 |
| Adjusted hazard ratio (95% CI), model 2 | 1.00 (ref) | 0.79 (0.63–0.99) | 0.04 | 1.00 (ref) | 1.12 (0.57–2.18) | 0.74 |
| Adjusted hazard ratio (95% CI), model 3 | 1.00 (ref) | 0.79 (0.63–0.99) | 0.04 | 1.00 (ref) | 1.14 (0.56–2.32) | 0.72 |
| Non‐cardiovascular death | ||||||
| Number of events | 173 | 320 | 18 | 41 | ||
| Event rate (per 1000 person‐year) | 19.3 | 16.7 | 14.5 | 17.1 | ||
| Unadjusted hazard ratio (95% CI) | 1.00 (ref) | 0.85 (0.71–1.02) | 0.08 | 1.00 (ref) | 1.19 (0.68–2.07) | 0.54 |
| Adjusted hazard ratio (95% CI), model 1 | 1.00 (ref) | 0.82 (0.68–0.99) | 0.04 | 1.00 (ref) | 1.46 (0.82–2.62) | 0.20 |
| Adjusted hazard ratio (95% CI), model 2 | 1.00 (ref) | 0.81 (0.67–0.97) | 0.02 | 1.00 (ref) | 1.71 (0.94–3.10) | 0.07 |
| Adjusted hazard ratio (95% CI), model 3 | 1.00 (ref) | 0.82 (0.68–0.99) | 0.04 | 1.00 (ref) | 1.47 (0.79–2.72) | 0.22 |
| Major adverse cardiovascular events | ||||||
| Number of events | 278 | 575 | 39 | 68 | ||
| Event rate (per 1000 person‐year) | 39.4 | 37.0 | 44.3 | 34.6 | ||
| Unadjusted hazard ratio (95% CI) | 1.00 (ref) | 0.94 (0.81–1.08) | 0.37 | 1.00 (ref) | 0.77 (0.52–1.14) | 0.19 |
| Adjusted hazard ratio (95% CI), model 1 | 1.00 (ref) | 0.90 (0.78–1.04) | 0.17 | 1.00 (ref) | 0.79 (0.52–1.18) | 0.24 |
| Adjusted hazard ratio (95% CI), model 2 | 1.00 (ref) | 0.92 (0.80–1.07) | 0.28 | 1.00 (ref) | 0.80 (0.52–1.21) | 0.28 |
| Adjusted hazard ratio (95% CI), model 3 | 1.00 (ref) | 0.93 (0.80–1.08) | 0.33 | 1.00 (ref) | 0.81 (0.52–1.25) | 0.33 |
| In patients with history of major CVD b | (n = 990) | (n = 2071) | (n = 131) | (n = 255) | ||
| All‐cause death | ||||||
| Number of events | 283 | 518 | 31 | 67 | ||
| Event rate (per 1000 person‐years) | 33.3 | 28.2 | 25.7 | 29.1 | ||
| Unadjusted hazard ratio (95% CI) | 1.00 (ref) | 0.83 (0.72–0.96) | 0.01 | 1.00 (ref) | 1.14 (0.75–1.75) | 0.54 |
| Adjusted hazard ratio (95% CI), model 1 | 1.00 (ref) | 0.80 (0.69–0.93) | 0.002 | 1.00 (ref) | 1.28 (0.82–2.00) | 0.28 |
| Adjusted hazard ratio (95% CI), model 2 | 1.00 (ref) | 0.79 (0.68–0.92) | 0.002 | 1.00 (ref) | 1.34 (0.85–2.13) | 0.21 |
| Adjusted hazard ratio (95% CI), model 3 | 1.00 (ref) | 0.79 (0.68–0.92) | 0.002 | 1.00 (ref) | 1.31 (0.81–2.11) | 0.26 |
| Cardiovascular death | ||||||
| Number of events | 119 | 217 | 14 | 28 | ||
| Event rate (per 1000 person‐year) | 14.0 | 11.8 | 11.6 | 12.2 | ||
| Unadjusted hazard ratio (95% CI) | 1.00 (ref) | 0.84 (0.67–1.05) | 0.11 | 1.00 (ref) | 1.05 (0.55–2.00) | 0.87 |
| Adjusted hazard ratio (95% CI), model 1 | 1.00 (ref) | 0.79 (0.63–0.99) | 0.03 | 1.00 (ref) | 1.11 (0.57–2.16) | 0.75 |
| Adjusted hazard ratio (95% CI), model 2 | 1.00 (ref) | 0.79 (0.63–0.99) | 0.04 | 1.00 (ref) | 1.04 (0.53–2.05) | 0.91 |
| Adjusted hazard ratio (95% CI), model 3 | 1.00 (ref) | 0.79 (0.62–0.99) | 0.04 | 1.00 (ref) | 1.09 (0.52–2.28) | 0.81 |
| Non‐cardiovascular death | ||||||
| Number of events | 164 | 301 | 17 | 39 | ||
| Event rate (per 1000 person‐year) | 19.3 | 16.4 | 14.1 | 17.0 | ||
| Unadjusted hazard ratio (95% CI) | 1.00 (ref) | 0.83 (0.69–1.00) | 0.05 | 1.00 (ref) | 1.22 (0.69–2.15) | 0.50 |
| Adjusted hazard ratio (95% CI), model 1 | 1.00 (ref) | 0.81 (0.67–0.98) | 0.03 | 1.00 (ref) | 1.50 (0.81–2.77) | 0.19 |
| Adjusted hazard ratio (95% CI), model 2 | 1.00 (ref) | 0.79 (0.65–0.96) | 0.01 | 1.00 (ref) | 1.70 (0.91–3.16) | 0.09 |
| Adjusted hazard ratio (95% CI), model 3 | 1.00 (ref) | 0.80 (0.65–0.97) | 0.02 | 1.00 (ref) | 1.41 (0.74–2.72) | 0.29 |
| Major adverse cardiovascular events | ||||||
| Number of events | 267 | 558 | 39 | 65 | ||
| Event rate (per 1000 person‐year) | 39.8 | 37.3 | 45.6 | 34.6 | ||
| Unadjusted hazard ratio (95% CI) | 1.00 (ref) | 0.93 (0.81–1.08) | 0.36 | 1.00 (ref) | 0.75 (0.50–1.11) | 0.15 |
| Adjusted hazard ratio (95% CI), model 1 | 1.00 (ref) | 0.90 (0.78–1.04) | 0.16 | 1.00 (ref) | 0.74 (0.49–1.12) | 0.15 |
| Adjusted hazard ratio (95% CI), model 2 | 1.00 (ref) | 0.92 (0.79–1.07) | 0.27 | 1.00 (ref) | 0.75 (0.49–1.14) | 0.17 |
| Adjusted hazard ratio (95% CI), model 3 | 1.00 (ref) | 0.92 (0.79–1.07) | 0.30 | 1.00 (ref) | 0.74 (0.47–1.15) | 0.17 |
Data are presented as number or hazard ratio (95% CI). Model 1 included age, sex, race/ethnicity, educational level, smoking status, BMI, duration of diabetes, hypertension and dyslipidaemia. Model 2 included the potential confounders of model 1 plus glycated haemoglobin, low‐density lipoprotein cholesterol, systolic blood pressure and the assigned glycaemic strategy of the ACCORD trial. Model 3 included all potential confounders of models 1 and 2 plus the use of certain types of medications (insulin, metformin, thiazolidinedione, angiotensin‐converting enzyme inhibitors/angiotensin II receptor blockers, calcium channel blockers, thiazide, beta‐blockers, statin and antiplatelet agents except aspirin), high‐density lipoprotein cholesterol, log‐transformed triglyceride, diastolic blood pressure, estimated glomerular filtration rate, and the assigned blood pressure and lipid strategy of the ACCORD trial.
Abbreviations: CI, confidence interval; CVD, cardiovascular disease.
Clinical cardiovascular disease was defined as myocardial infarction, angina pectoris, coronary revascularization including coronary artery bypass grafting and percutaneous coronary intervention, stroke or other revascularization such as carotid artery revascularization and peripheral artery revascularization.
Major cardiovascular disease was defined as coronary heart disease or stroke.
Similarly, the risks of cardiovascular and non‐cardiovascular mortality in patients not using NSAIDs were significantly lower in those using aspirin than in those not using aspirin, whereas these risks in patients using NSAIDs did not differ significantly between the two groups (Figure S3 and Table 2).The analyses limited to patients with major cardiovascular diseases or coronary heart disease showed similar results. Regarding MACE (Figure S4), no significant interactions were noted between the use of aspirin and use of NSAIDs.
4. DISCUSSION
Aspirin acts by irreversibly acetylating the serine residue at position 529 in platelet COX‐1 and inhibits thromboxane production by platelets,6 which results in inhibition of platelet aggregation and a lower risk of cardiovascular events, particularly in patients with a history of cardiovascular disease.1 Some,10 but not all,11 studies revealed that the associations between the use of aspirin and cardiovascular events may be affected by the use of NSAIDs. One possible explanation is that NSAIDs block the access of aspirin to the acetylation site of platelet‐expressed COX‐1.6, 7 However, the clinical data are inconsistent. Kurth et al. previously described a subgroup analysis of a 5‐year randomized, double‐blind, placebo‐controlled trial concerning the use of 325 mg aspirin on alternate days among 22 071 apparently healthy US male physicians for whom prospective observational data on the use of NSAIDs were available.12 The analyses suggested that regular, but not intermittent, use of NSAIDs reduced the clinical benefits of aspirin use in avoiding a first myocardial infarction. Conversely, Patel et al. demonstrated a 40% reduction in the rate of development of myocardial infarction in patients using both aspirin and ibuprofen as opposed to using aspirin alone, suggesting enhanced protection against acute myocardial infarction with co‐administration of aspirin and ibuprofen.11 In addition, no study evaluated whether NSAIDs interfere with the well‐established beneficial effects of aspirin in diabetic patients with a history of cardiovascular disease. Specifically considering this patient group, the present study revealed that aspirin use was associated with lower mortality only in those not using NSAIDs; in patients using NSAIDs, no beneficial effect of aspirin was identified. These results suggest that the effects of aspirin are attenuated by use of NSAIDs. In diabetic patients with a history of cardiovascular disease, our findings indicate that the best choice of treatment would be to minimize the use of NSAIDs. Notably, the present study also demonstrated that the use of aspirin was associated with a decreased risk of non‐cardiovascular mortality that was attenuated by concomitant use of NSAIDs. Several studies have suggested that aspirin use reduces the incidence of cancer and of non‐cardiovascular and cancer mortality.13 Some studies using animal models suggested that these effects are mediated, in part, by inhibition of the COX enzymes and reduced production of inflammatory mediators.13 NSAIDs have been associated with gastrointestinal events and renal failure, and concomitant use of aspirin and NSAIDs can in fact lead to an increased risk of gastrointestinal events relative to the use of NSAIDs alone.14 These adverse effects of NSAIDs may also reduce the beneficial effects of aspirin. The effect of aspirin on MACE was not significantly affected by the use of NSAIDs in this study. Further, the incidence of MACE tended to be lower in patients using NSAIDs and aspirin than in those using either NSAIDs or aspirin alone. Therefore, further large‐scale studies are warranted to validate the results of this study.
This study has several limitations. First, this was not a randomized trial but an observational study. Even after multivariable adjustments, residual bias, including unknown confounders and unmeasured underlying disease with chronic pain, such as rheumatoid arthritis, might have affected the results. Second, the small number of events and of patients using NSAIDs might influence the results. In addition, we could not completely assess the interactions between use of aspirin and nonselective or COX‐2‐selective NSAIDs (eg, celecoxib vs. naproxen) and the differential impact on results. Hence, our results should be confirmed by larger‐scale studies with detailed information concerning the use of NSAIDs. Third, the management of diabetes has changed since the ACCORD study was initiated. The residual risk is decreasing because of additional effective therapies, such as increased use of statins and sodium glucose cotransporter 2 inhibitors. Thus, it remains unknown whether the same results would be obtained in a more contemporary cohort. Fourth, detailed information regarding the doses of aspirin and the doses and types of NSAIDs is lacking. A recent study revealed that low doses (75–100 mg) of aspirin were effective in preventing cardiovascular events only in patients weighing less than 70 kg, whereas high doses (≥325 mg) of aspirin were effective only in patients weighing 70 kg or more.15 Those findings were consistent in diabetic patients. In addition, the pharmacodynamic interactions that occur in platelets may differ according to the different types of NSAIDs involved.6, 10 Moreover, the timing of aspirin and NSAIDs use may also influence the results.6 Further studies with more detailed information about NSAIDs are needed to reveal the interactions between aspirin and NSAIDs.
In conclusion, the present study demonstrated that, in type 2 diabetic patients with a history of cardiovascular disease, aspirin use is beneficial for those who are not using NSAIDs but is not beneficial for those who are using NSAIDs. To fully derive the benefits of aspirin and avoid the increased risk of cardiovascular events, more attention should be paid to the use of NSAIDs.
CONFLICT OF INTEREST
The authors have no conflicts of interest to report.
AUTHOR CONTRIBUTIONS
T. T. was responsible for the study concept and design, for data acquisition and for statistical analysis. T. T. and H. K. shared responsibility for analysis and interpretation of data. T. T. drafted the manuscript.
Supporting information
Figure S1. Kaplan–Meier survival curves for all‐cause mortality in patients using and not using aspirin
Figure S2. Kaplan–Meier survival curves for all‐cause mortality in propensity score‐matched patients using and not using aspirin
Figure S3. Kaplan–Meier survival curves for cardiovascular and non‐cardiovascular mortality in patients using and not using aspirin
Figure S4. Kaplan–Meier survival curves for major adverse cardiovascular events in patients using and not using aspirin
Table S1. Baseline characteristics of propensity score‐matched patients with type 2 diabetes and a history of clinical cardiovascular disease using and not using nonsteroidal anti‐inflammatory drugs
ACKNOWLEDGMENTS
This manuscript was prepared using ACCORD and ACCORDION research materials obtained from the NHLBI Biologic Specimen and Data Repository Information Coordinating Center and does not necessarily reflect the opinions or views of the ACCORD/ACCORDION or the NHLBI.
Tsujimoto T, Kajio H. No beneficial effects of aspirin on secondary cardiovascular prevention in patients with type 2 diabetes using non‐steroidal anti‐inflammatory drugs. Diabetes Obes Metab. 2019;21:1978–1984. 10.1111/dom.13737
Funding information This study was supported by a grant from the National Center for Global Health and Medicine (Grant Number: 30‐1001) and a Grant‐in‐Aid for Scientific Research from the Japan Society for the Promotion of Science (Grant Number: 18K16219). These funding sources had no role in the design and conduct of the study; collection, management, analysis and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Peer Review The peer review history for this article is available at https://publons.com/publon/10.1111/dom.13737.
REFERENCES
- 1. Baigent C, Blackwell L, Collins R, et al. Aspirin in the primary and secondary prevention of vascular disease: collaborative meta‐analysis of individual participant data from randomised trials. Lancet. 2009;373:1849‐1860. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Bhala N, Emberson J, Merhi A, et al. Vascular and upper gastrointestinal effects of non‐steroidal anti‐inflammatory drugs: meta‐analyses of individual participant data from randomised trials. Lancet. 2013;382:769‐779. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. McGettigan P, Henry D. Cardiovascular risk and inhibition of cyclooxygenase: a systematic review of the observational studies of selective and nonselective inhibitors of cyclooxygenase 2. JAMA. 2006;296:1633‐1644. [DOI] [PubMed] [Google Scholar]
- 4. Bally M, Dendukuri N, Rich B, et al. Risk of acute myocardial infarction with NSAIDs in real world use: bayesian meta‐analysis of individual patient data. BMJ. 2017;357:j1909. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Gislason GH, Jacobsen S, Rasmussen JN, et al. Risk of death or reinfarction associated with the use of selective cyclooxygenase‐2 inhibitors and nonselective nonsteroidal antiinflammatory drugs after acute myocardial infarction. Circulation. 2006;113:2906‐2913. [DOI] [PubMed] [Google Scholar]
- 6. Catella‐Lawson F, Reilly MP, Kapoor SC, et al. Cyclooxygenase inhibitors and the antiplatelet effects of aspirin. N Engl J Med. 2001;345:1809‐1817. [DOI] [PubMed] [Google Scholar]
- 7. Rimon G, Sidhu RS, Lauver DA, et al. Coxibs interfere with the action of aspirin by binding tightly to one monomer of cyclooxygenase‐1. Proc Natl Acad Sci U S A. 2010;107:28‐33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Gerstein HC, Miller ME, Byington RP, et al. Effects of intensive glucose lowering in type 2 diabetes. N Engl J Med. 2008;358:2545‐2559. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. ACCORD Study Group . Nine‐year effects of 3.7 years of intensive glycemic control on cardiovascular outcomes. Diabetes Care. 2016;39:701‐708. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. MacDonald TM, Wei L. Effect of ibuprofen on cardioprotective effect of aspirin. Lancet. 2003;361:573‐574. [DOI] [PubMed] [Google Scholar]
- 11. Patel TN, Goldberg KC. Use of aspirin and ibuprofen compared with aspirin alone and the risk of myocardial infarction. Arch Intern Med. 2004;164:852‐856. [DOI] [PubMed] [Google Scholar]
- 12. Kurth T, Glynn RJ, Walker AM, et al. Inhibition of clinical benefits of aspirin on first myocardial infarction by nonsteroidal antiinflammatory drugs. Circulation. 2003;108:1191‐1195. [DOI] [PubMed] [Google Scholar]
- 13. Rothwell PM, Fowkes FG, Belch JF, Ogawa H, Warlow CP, Meade TW. Effect of daily aspirin on long‐term risk of death due to cancer: analysis of individual patient data from randomised trials. Lancet. 2011;377:31‐41. [DOI] [PubMed] [Google Scholar]
- 14. Hippisley‐Cox J, Coupland C, Logan R. Risk of adverse gastrointestinal outcomes in patients taking cyclo‐oxygenase‐2 inhibitors or conventional non‐steroidal anti‐inflammatory drugs: population based nested case‐control analysis. BMJ. 2005;331:1310‐1316. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Rothwell PM, Cook NR, Gaziano JM, et al. Effects of aspirin on risks of vascular events and cancer according to bodyweight and dose: analysis of individual patient data from randomised trials. Lancet. 2018;392:387‐399. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1. Kaplan–Meier survival curves for all‐cause mortality in patients using and not using aspirin
Figure S2. Kaplan–Meier survival curves for all‐cause mortality in propensity score‐matched patients using and not using aspirin
Figure S3. Kaplan–Meier survival curves for cardiovascular and non‐cardiovascular mortality in patients using and not using aspirin
Figure S4. Kaplan–Meier survival curves for major adverse cardiovascular events in patients using and not using aspirin
Table S1. Baseline characteristics of propensity score‐matched patients with type 2 diabetes and a history of clinical cardiovascular disease using and not using nonsteroidal anti‐inflammatory drugs