EClinicalMedicine 27 (2020) 100548 Contents lists available at ScienceDirect EClinicalMedicine journal homepage: https://www.journals.elsevier.com/eclinicalmedicine Research Paper Aspirin for primary prevention of ST segment elevation myocardial infarction in persons with diabetes and multiple risk factors Raffaele Bugiardinia,*, Sasa Pavasovica,b, Jinsung Yoonc, Mihaela van der Schaard, Sasko Kedeve, Marija Vavlukise, Zorana Vasiljevicf, Maria Bergamia, Davor Milicicb, Olivia Manfrinia, Edina Cenkoa, Lina Badimong a Department of Experimental, Diagnostic and Specialty Medicine, University of Bologna, Policlinico Sant’Orsola Malpighi, Padiglione 11, Via Massarenti 9, 40138 Bologna, Italy b Department for Cardiovascular Diseases, University Hospital Centre Zagreb, University of Zagreb, Zagreb, Croatia c Google Cloud AI, Sunnyvale, CA, USA d Cambridge Centre for Artificial Intelligence in Medicine, Department of Applied Mathematics and Theoretical Physics and Department of Population Health, Uni- versity of Cambridge, Cambridge, United Kingdom e University Clinic of Cardiology, Medical Faculty, University "Ss. Cyril and Methodius", Skopje, Macedonia f Medical Faculty, University of Belgrade, Belgrade, Serbia g Cardiovascular Research Program ICCC, IR-IIB Sant Pau, Hospital de la Santa Creu i Sant Pau, CiberCV-Institute Carlos III, Barcelona, Spain A R T I C L E I N F O A B S T R A C T Article History: Background: Controversy exists as to whether low-dose aspirin use may give benefit in primary prevention of Received 20 June 2020 cardiovascular (CV) events. We hypothesized that the benefits of aspirin are underevaluated. Revised 12 August 2020 Methods: We investigated 12,123 Caucasian patients presenting to hospital with acute coronary syndromes Accepted 28 August 2020 as first manifestation of CV disease from 2010 to 2019 in the ISACS-TC multicenter registry (ClinicalTrials. Available online 20 September 2020 gov, NCT01218776). Individual risk of ST segment elevation myocardial infarction (STEMI) and its association with 30-day mortality was quantified using inverse probability of treatment weighting models matching for Keywords: concomitant medications. Estimates were compared by test of interaction on the log scale. Aspirin Primary prevention Findings: The risk of STEMI was lower in the aspirin users (absolute reduction: 6¢8%; OR: 0¢73; 95%CI: ST segment elevation myocardial infarction 0¢650¢82) regardless of sex (p for interaction=0¢1962) or age (p for interaction=0¢1209). Benefits of aspirin were seen in patients with hypertension, hypercholesterolemia, and in smokers. In contrast, aspirin failed to demonstrate a significant risk reduction in STEMI among diabetic patients (OR:1¢10;95%CI:0¢891¢35) with a significant interaction (p: <0¢0001) when compared with controls (OR:0¢64,95%CI:0¢560¢73). Stratification of diabetes in risk categories revealed benefits (p interaction=0¢0864) only in patients with concomitant hypertension and hypercholesterolemia (OR:0¢87, 95% CI:0¢651¢15), but not in smokers. STEMI was strongly related to 30-day mortality (OR:1¢93; 95%CI:1¢592¢35) Interpretation: Low-dose aspirin reduces the risk of STEMI as initial manifestation of CV disease with potential benefit in mortality. Patients with diabetes derive substantial benefit from aspirin only in the presence of multiple risk factors. In the era of precision medicine, a more tailored strategy is required. Funding: None. © 2020 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/) 1. Introduction 6.6 million do so without a health care provider recommendation [1]. All of these patients endorse the aspirin use despite the Aspirin is still on the medication list of a new patient with announcement in August 2018 about two clinical trials that hypertension and hypercholesterolemia and no history of clini- found aspirin offered few benefits for healthy adults and might cally evident cardiovascular (CV) disease. According to the even raise their risk of bleeding, namely the ARRIVE (Aspirin to National Health Interview Survey, about 29 million patients who Reduce Risk of Initial Vascular Events) [2] and the ASCEND (A do not have CV disease take aspirin daily for prevention—and Study of Cardiovascular Events in Diabetes) [3]. Notably, a large proportion of patients in both studies were taking statins and antihypertensive drugs, and only a small proportion were current * Corresponding author. smokers. E-mail address: raffaele.bugiardini@unibo.it (R. Bugiardini). https://doi.org/10.1016/j.eclinm.2020.100548 2589-5370/© 2020 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/) 2 R. Bugiardini et al. / EClinicalMedicine 27 (2020) 100548 CHD from STEMI to non-ST elevation acute coronary syndromes Research in context (NSTE-ACS) or to stable angina, may reduce the overall patient risk and ultimately CV mortality. Evidence before this study The second question is whether concomitant preventive medica- The role of low-dose aspirin in the general population without tions may blunt the cardio-protective effect of aspirin. Currently, many CV disease remains controversial. The USPSTF 2016 guidelines patients are intensively treated with beta-blockers, angiotensin convert- recommended low-dose aspirin to high-risk individuals aged ing enzyme (ACE) inhibitors or angiotensin receptor blockers (ARBs) 5069 years with 10year cardiovascular risk 10%, regardless and statins. Widespread statin use is the norm in clinical practice. Their of the presence or absence of diabetes. The 2019 Scientific role in prevention therapy is still unsettled. Concomitant medications Statement by the American Diabetes Association recommended may be tracked as both a history item, as well as, during the study. Lack aspirin therapy for primary prevention for diabetic patients of consistency on concomitant medications reporting makes it difficult with 10-year CV risk 10% provided that these patients are to correlate the data with the primary endpoints of the study. While aged more than 50 and less than 70 years and have at least one this issue is clearly important, there is no or little information on con- additional major risk factor. No previous study has estimated comitant medications in prior aspirin prevention trials. the effects of low-dose aspirin on prevention of ST segment ele- A further question is whether aspirin may have a margin of bene- vation myocardial infarction as initial clinical presentation of fit in specific populations such as healthy older adults [5] or in pres- coronary heart disease in individuals with diabetes alone or ence of specific comorbid conditions such as diabetes [3]. In addition, diabetes in combination with other conventional risk factors. some trials have documented sex-specific heterogeneity of treatment effects [6]. Careful analysis of the data does not allow one to draw Added value of this study firm conclusions on these points. Heterogeneity in the predicted out- come risk deserves further investigations. To our knowledge, this study provides the first estimate of ben- The most commonly used method of examining whether treatment efit of aspirin prophylaxis in patients with diabetes depending effects vary in a trial population is to serially divide patients into sub- on the number and type of associated risk factors. No beneficial groups based on potentially relevant characteristics. In the current era, effect of aspirin was seen in individuals with diabetes alone the main problem with this conventional approach is that better man- and in those who had one additional major conventional risk agement of risk factors has lowered the risk of developing disease and factor. Low-dose aspirin was insufficient to confer protection in disease severity. Hence, the number of patients that must be enrolled diabetic patients who were smokers. Conversely, prevention of in a trial becomes much greater as lower risk patients are included. ST segment elevation myocardial infarction through aspirin use Thus, subgroup analyses may be underpowered. On the other hand, a was consistent in diabetic individuals who were concomitantly randomized trial cannot be done for every subgroup. One approach affected by hypertension and hypercholesterolemia. that can help to counter this problem is to carry out an aspirin preven- tion study using a register-based cohort data in a case-control or Implications of all the available evidence match weighted design. In such type of studies, groups are defined by Implications of all the available evidence: According to the the outcome. Researchers then look back to ascertain each person's National Health Interview Survey, about 29 million patients exposure status to aspirin, and to compare the frequency of outcome who do not have clinically CV disease take aspirin daily for pre- in the aspirin group with that in the control group. Investigators would vention, and 6¢6 million do so without a health care provider correctly address confounding due to concomitant medications and recommendation. evaluate the potential incremental value of preventing cardiovascular In 2016, about 1¢7 million hospital discharges for major car- outcomes by using aspirin on the top of these medications. diovascular diseases were reported with diabetes as any listed The current investigation reflects this approach. All patients had a diagnosis among US adults including 438,000 for ischemic heart first clinical presentation of CHD documented by a diagnosis of acute disease. coronary syndromes. The outcome of interest was STEMI, since it is a Clinical and public health efforts should focus on identifying marker of poorer short-term prognosis. A requirement of the study optimal preventive measures for the whole diabetic population was that the outcomes of patients exposed to aspirin could not be and individual patients. influenced by other potential prevention therapies. This task was achieved by matching concomitant medications using inverse proba- bility of treatment weighting. Aspirin users versus nonusers had a These new studies push the pendulum away from aspirin prophy- similar pattern of exposure to concomitant medications. laxis for primary prevention. However, in our view, differences between benefits and harms are likely to be a razor thin, as balancing one adverse 2. Methods CV event against one bleeding event is not straightforward. For exam- ple, an episode of ST segment elevation myocardial infarction (STEMI), 2.1. Derivation cohort is more lethal than a minimally symptomatic gastrointestinal bleeding, whereas some bleeding events, such as severe intracranial haemor- The International Survey of Acute Coronary Syndromes in Transi- rhages, are more lethal than some ischemic CV events, such as episodes tional Countries (ISACS-TC; NCT01218776) is a large observational of stable or unstable angina. Prospectively, we cannot predict which of and multinational registry. Methods have been previously described those outcomes would apply to any given patient and to any given sex. [7,8]. Data were collected from 41 centres in 12 countries. The Uni- The role of aspirin in primary prevention has, therefore, become increas- versity of Bologna is the data-coordinating centre and is responsible ingly uncertain and many important questions remain unsolved. for the budget and the quality of data delivered to the ISACS-TC. Since The first question is whether aspirin is reducing the incidence of information was collected anonymously, institutional review boards STEMI, which is the most serious and catastrophic clinical manifesta- waived the need for individual informed consent. tion of coronary heart disease (CHD). Relying on the generic outcome of “myocardial infarction” to ascertain a benefit from aspirin use 2.2. Patient population might be challenging, as reduction in short-term case fatality rates for myocardial infarction in US appears to be driven by a decreased The initial population consisted of 20,189 patients with ACS incidence of STEMI [4]. Shifting the mode of initial presentations of enrolled between January 2010 and January 2019. Patients R. Bugiardini et al. / EClinicalMedicine 27 (2020) 100548 3 Fig. 1. Estimated effects of aspirin on STEMI: distribution by CV risk factors. Association between use of aspirin before index event and incidence of ST elevation myocardial infarc- tion sorted by the presence of one traditional risk factor. Abbreviations: OR, Odds Ratio; CI, Confidence Interval; CV, cardiovascular; STEMI, ST elevation myocardial infarction. presenting with a history of a vascular events were excluded. Patients continuous variables. A 2-sided p value of <0¢05 was considered statis- using thienopyridine before index admission were also excluded. The tically significant. Variables describing demographic characteristics, final population consisted of 12,123 patients (60.0% of the overall medical history, cardiovascular risk factors, and clinical features at hos- study population) (Fig. 1 in the Supplement). pital presentation are reported in Supplemental Table 1. We used inverse probability of treatment weighting and logistic regression 2.3. Main outcome measures and definitions models to assess the effect of variables on the associations of interest (Supplemental Methods, page 4) [10]. Standardized differences after The primary outcome measure was the rate of STEMI at hospital weighting were calculated to ensure balanced treatment groups with presentation. STEMI is a predictor of heart failure and 30-day mortal- respect to baseline characteristics. Groups were considered balanced ity. In the current study, heart failure was defined as Killip class 2. when the standardized difference was less than 20% (Supplemental We defined prior low-dose (75/150 mg/day) aspirin, statins, ACE Methods, page 4). We calculated ORs with their 95% CIs from these inhibitors, ARBs and beta-blockers users as those patients who had models. Fixed covariates included demographic information and base- taken these medications on a regular basis at least for two weeks line clinical characteristics (Table 1). We had complete data on aspirin before the onset of the qualifying event. Medications received imme- use, sex, age and index event. Some patients had missing data on other diately before hospitalization or in the emergency department were variables. We used k-nearest neighbour (KNN) algorithms as imputa- not considered prior medication use. We defined multivessel disease tion method to treat missing data [11,12]. (Supplemental Methods, as at least two main branches of the epicardial coronary artery with page 5). Separate analyses were done to quantify the specific impact of 70% or more stenotic lesions, or a 50% or greater stenosis in the left aspirin use on STEMI rate for each of the traditional risk factors. For main coronary artery. Smoking habits, weight and height were self- these analyses, we divided the risk factors into dichotomous variables reported. The specific risks of STEMI were estimated for current and and grouped aspirin users and nonusers in those with and without the former smokers compared with never smokers. For simplicity, the risk factor under consideration. We estimated the odds ratios (ORs) categories of current and former smoking were collapsed in some and 95% confidence intervals (CI) in patients with and without the risk analyses, specifically on the effect of smoking in association with factor under consideration. Estimates were compared by test of inter- other risk factors. Hypertension, hypercholesterolemia and diabetes action on the log scale (Supplemental Methods, page 6) [13]. were assessed by designation of medical history prior to admission in the database. Body mass index (BMI) was calculated as weight (kg) 2.5. Role of the funding source divided by height squared (m2). (Supplemental Methods, page 4). The 10-year CV risk for each patient was calculated by using the No sponsor had any role in the design of the study or in the collec- Pooled Cohort Equations. We set the cut-off for increased level of CV tion, analysis, interpretation of data, in the writing of the report, and/ disease risk at 10% according to the 2016 final recommendation or in the decision to submit the paper for publication. The Principal statement of the USPS Task Force [9]. Investigator had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the 2.4. Statistical analysis data analysis. Baseline characteristics were reported as percentages for categori- 3. Results cal variables and means with standard deviation (SD) for continuous variables. Comparisons between groups were made either by Pearson Overall, 12,123 patients entered into the study. (Supplemental Fig. 1). x2 test for baseline categorical variables or two-sample t-test for Patient baseline characteristics are listed in Supplemental Table 1. 4 R. Bugiardini et al. / EClinicalMedicine 27 (2020) 100548 Table 1 Inverse probability of treatment weighting: outcomes sorted by aspirin use before index event. Characteristics Overall population Aspirin usersN = 1506 Aspirin nonusersN = 10,617 Standardized difference Age, y 62¢7 § 12¢5 61¢4 § 12¢2 0¢1046 Cardiovascular risk factors Diabetes 25¢4 22¢6 0¢0661 History of hypertension 69¢8 65¢4 0¢0955 History of hypercholesterolemia 43¢0 38¢8 0¢0870 Current smokers 43¢6 44¢5 0¢0164 Former smokers 8¢4 7¢4 0¢0372 Clinical history COPD 6¢0 5¢2 0¢0367 Chronic kidney disease 6¢4 5¢1 0¢0599 Medications before admission Statins 10¢5 9¢3 0¢0413 ACE inhibitors/ ARBs 38¢8 33¢2 0¢1174 Beta blockers 24¢1 20¢1 0¢0962 Angiographic findings Multivessel disease 44¢2 42¢5 0¢0338 Outcome STEMI 65¢4 72¢2 0¢1459* Odds Ratio (95%CI) 0¢73 (0¢65  0¢82) 0¢1459** Data are percentages or means § Standard deviation (SD) unless stated otherwise. *P-value for STEMI for aspirin users versus non-users <0¢0001; ** P-value <0¢0001. Abbreviations: ACE, angiotensin converting enzyme; ARBs, angiotensin receptor blockers; COPD, chronic obstructive pul- monary disease; STEMI, ST elevation myocardial infarction. The distribution of diabetic patients in those taking oral antidia- presence or absence of the four conventional risk factors (hyperten- betics, insulin or controlling the disease with diet is shown in sion, hypercholesterolemia, diabetes mellitus, and smoking) (Fig. 1). Supplemental Fig. 2. Slightly more than 12% (n = 1506) of patients In all subgroups apart from diabetes, the benefits of aspirin were reported use of aspirin. Aspirin users were older. They also had approximately similar (Supplemental Tables 710) to those seen in more frequently diabetes, hypertension, hypercholesterolemia, the overall patient population. Aspirin use was associated with a sub- and chronic kidney disease compared with aspirin nonusers. stantially lower risk of presenting with STEMI in patients who were Aspirin users were also more likely to take concomitant evi- current (OR: 0¢78; 95% CI 0¢640¢96) or former smokers (OR: 0¢63; dence-based medications. A detailed description of the sociode- 95% CI 0¢44 to 0¢90), for those with hypercholesterolemia (OR: 0¢73; mographic and clinical characteristics associated with 10-year CV 95% CI 0¢620¢87), and for those with hypertension (OR: 0¢72; 95% CI risk categories is presented in Supplemental Table 2. Aspirin users 0¢640¢82). Conversely, in diabetic patients (Supplemental Table 11), showed higher predicted 10-year CV risk compared with nonus- aspirin treatment was associated with a nonsignificant increase in ers. Clinical presentation with STEMI as index event was strongly the main end point of STEMI (OR: 1¢10, 95% CI: 0¢891¢35). Analysis related to 30-day mortality (OR: 1¢93; 95%CI: 1¢592¢35) and of the overall population of patients excluding patients with diabetes heart failure on hospital admission (Supplemental Figs. 3 and 4). showed opposite results: (OR: 0¢64;95% CI: 0¢560¢73). The interac- tion between the two subgroups was highly significant (p < 0¢0001) 3.1. Balancing covariates and aspirin use in the overall population (Fig. 1). Inverse probability of treatment weighting eliminated most of the 3.3. Diabetes and one more risk factor differences between covariates of aspirin users versus nonusers (Table 1). Baseline characteristics were balanced between the two We compared outcomes in aspirin users versus nonusers in study groups for patient demographics, disease risk factors and prior patients having two conventional risk factors of which one was dia- evidence-based medication use. Prior aspirin use was associated with betes (Fig. 2). For these analyses we combined current and former a significantly decreased rate of STEMI as compared with no prior smokers in a single category. In patients who were smokers, (Supple- aspirin use (absolute difference 6¢8%; OR, 0¢73; 95%CI 0¢650¢82). mental Table 12), aspirin was associated with a nonsignificant 25% Next, we investigated the association of prior aspirin use with out- increase in STEMI (OR: 1¢25; 95% CI, 0¢871¢80). There was power to comes separately in women versus men, in patients aged 50 but less detect interaction with controls (interaction: p = 0¢0010). In patients than 60 versus 6069 years, and in patients with BMI <25 kg/m2 ver- with history of hypercholesterolemia (Supplemental Table 13), aspi- sus those with BMI 25 kg/m2. Effects were consistent for all sub- rin was associated with a nonsignificant 8% reduction of STEMI (OR: groups (Tables 24), with no interaction effects by age, sex or 0¢92; 95% CI, 0¢701¢22). The relative risk compared with that of con- weight. (Supplemental Tables 35). Benefits of aspirin were also trols was significantly different (interaction: p = 0¢039). In patients observed when the analysis was restricted to the higher CV risk with history of hypertension (Supplemental Table 14), aspirin use group (absolute difference 9¢0%; OR, 0¢67; 95%CI 0¢590¢75), but resulted in a not statistically significant 4% increase of STEMI (OR were not seen in the lower risk group (absolute difference 2¢4%; OR ratio: 1¢04; 95% CI, 0¢831¢29). Again, there was power to detect ratio, 0¢88; 95%CI 0¢661¢19). (Table 5). The relative risks from the interaction with controls (p = 0¢0004). CV risk subgroups significantly differed from each other (Supplemen- tal Table 6) 3.4. Diabetes and multiple risk factors 3.2. Subgroup analyses sorted by specific risk factors Diabetic patients were separated into strata based on the type of association with two or more conventional risk factors (Fig. 3). Com- To explore potential sources of heterogeneity, we evaluated the pared with controls, diabetic patients with two more risk factors did primary end point of STEMI in subgroups of patients defined by the not benefit of aspirin when diabetes was accompanied by smoking, R. Bugiardini et al. / EClinicalMedicine 27 (2020) 100548 5 Table 2 Inverse probability of treatment weighting: outcomes sorted by sex and aspirin use before index event. Characteristics Women Men Aspirin users Aspirin nonusers Standardized Aspirin users Aspirin nonusers Standardized N = 563 N = 3112 difference N = 943 N = 7505 difference Age, y 66¢6 § 11¢6 65¢6 § 12¢2 0¢0918 60¢8 § 12¢5 59¢6 § 11¢8 0¢0996 Cardiovascular risk factors Diabetes 28¢6 27¢6 0¢0242 24¢1 20¢5 0¢0875 History of hypertension 74¢6 73¢9 0¢0158 67¢2 61¢7 0¢1152 History of 44¢1 40¢4 0¢0758 42¢3 38¢1 0¢0856 hypercholesterolemia Current smokers 34¢4 31¢7 0¢0556 47¢7 50¢0 0¢0454 Former smokers 4¢4 3¢9 0¢0271 10¢5 9¢0 0¢0511 Clinical history COPD 5¢7 5¢7 0¢0022 6¢2 5¢0 0¢0556 Chronic kidney disease 7¢4 6¢0 0¢0572 6¢0 4¢7 0¢0576 Medications before admission Statins 12¢4 11¢2 0¢0380 9¢4 8¢4 0¢0344 ACE inhibitors/ ARBs 46¢7 42¢1 0¢0930 35¢1 29¢3 0¢1241 Beta blockers 31¢0 26¢8 0¢0920 20¢9 17¢3 0¢0923 Angiographic findings Multivessel disease 44¢3 41¢6 0¢0545 44¢0 43¢0 0¢0205 Outcome STEMI 64¢5 69¢7 0¢1103* 66¢0 73¢3 0¢1587x Odds Ratio (95% CI) 0¢79 (0¢65  0¢96) 0¢1103** 0¢71 (0¢61  0¢82) 0¢1587xx Data are percentages or means § Standard deviation (SD) unless stated otherwise. *P-value for STEMI for aspirin users versus non-users 0¢0179; ** P-value 0¢0148. x P-value for STEMI for aspirin users versus non-users <0¢0001. xx P-value <0¢0001. Abbreviations: ACE, angiotensin converting enzyme; ARBs, angiotensin receptor blockers; COPD, chronic obstructive pulmonary disease; STEMI, ST elevation myocardial infarction. Table 3 Inverse probability of treatment weighting: outcomes sorted by age and aspirin use before index event. Characteristics 5059 years 6069 years Aspirin users Aspirin nonusers Standardized difference Aspirin users Aspirin nonusers Standardized N = 297 N = 3066 N = 487 N = 3047 difference Age, y 54¢9 § 2¢7 54¢9 § 2¢9 0¢0042 64¢0 § 3¢0 64¢0 § 2¢9 0¢0133 Cardiovascular risk factors Diabetes 15¢9 18¢9 0¢0798 29¢0 26¢9 0¢0454 History of hypertension 64¢6 61¢5 0¢0641 74¢9 70¢2 0¢1051 History of 50¢3 41¢3 0¢1810 42¢6 41¢0 0¢0340 hypercholesterolemia Current smokers 59¢4 57¢6 0¢0347 44¢6 42¢5 0¢0423 Former smokers 8¢1 6¢7 0¢0536 10¢4 9¢0 0¢0454 Clinical history COPD 3¢2 3¢4 0¢0103 7¢8 6¢0 0¢0707 Chronic kidney disease 3¢4 2¢9 0¢0293 6¢0 5¢2 0¢0354 Medications before admission Statins 9¢9 8¢1 0¢0630 11¢9 10¢7 0¢0389 ACE inhibitors/ ARBs 35¢3 27¢3 0¢1958 42¢8 38¢6 0¢0858 Beta blockers 21¢1 16¢9 0¢1081 25¢7 22¢7 0¢0699 Angiographic findings Multivessel disease 38¢5 38¢0 0¢0108 47¢7 44¢2 0¢0688 Outcome STEMI 66¢3 72¢3 0¢1303* 64¢4 74¢4 0¢2200x Odds Ratio (95% CI) 0¢75 (0¢59  0¢97) 0¢1303** 0¢62 (0¢51  0¢76) 0¢2200xx Data are percentages or means § Standard deviation (SD) unless stated otherwise. *P-value for STEMI for aspirin users versus non-users 0¢0370; ** P-value 0¢0288. x P-value for STEMI for aspirin users versus non-users <0¢0001. xx P-value <0¢0001. Abbreviations: ACE, angiotensin converting enzyme; ARBs, angiotensin receptor blockers; COPD, chronic obstructive pulmonary disease; STEMI, ST ele- vation myocardial infarction. either in patients who additionally had history of hypercholesterol- only cluster of risk factors that improved ischemic outcomes with emia (OR ratio: 1¢25; 95% CI, 0¢801¢94; interaction test: p = 0¢0067) aspirin was the association of diabetes with history of hypercholes- or in those who had history of hypertension (OR ratio: 1¢11; 95% CI, terolemia and hypertension (Supplemental Table 18). In this high- 0¢771¢62; interaction test: p = 0¢0104) (Supplemental Tables 15 and risk population, the estimated treatment benefit was a 13% 16). Diabetic patients did not benefit of aspirin even combining all reduction in risk of STEMI, which was statistically insignificant (OR: four conventional risk factors into the model (OR: 1¢26; 95% CI, 0¢87; 95% CI, 0¢651¢15) compared with a 30% reduction in controls, 0¢801¢99; interaction test: p = 0¢0073) (Supplemental Table 17). The which was significant (OR: 0¢70; 95% CI, 0¢620¢80). But the relative 6 R. Bugiardini et al. / EClinicalMedicine 27 (2020) 100548 Table 4 Inverse probability of treatment weighting: outcomes sorted by BMI and aspirin use before index event. Characteristics BMI <25 BMI 25 Aspirin users Aspirin Standardized Aspirin users Aspirin nonusers Standardized N = 430 nonusersN = 3381 difference N = 1076 N = 7236 difference Age, y 63¢3 § 13¢2 63¢1 § 12¢5 0¢0163 62¢3 § 12¢2 60¢6 § 12¢0 0¢1445 Cardiovascular risk factors Diabetes 19¢7 16¢7 0¢0792 28¢0 25¢3 0¢0610 History of hypertension 60¢7 56¢9 0¢0772 74¢3 69¢3 0¢1133 History of 33¢6 30¢6 0¢0636 47¢2 42¢5 0¢0943 hypercholesterolemia Current smokers 47¢2 44¢8 0¢0480 42¢2 44¢3 0¢0418 Former smokers 7¢1 6¢0 0¢0430 9¢1 8¢0 0¢0394 Clinical history COPD 5¢2 4¢9 0¢0159 6¢2 5¢3 0¢0401 Chronic kidney disease 7¢4 5¢4 0¢0797 5¢7 4¢9 0¢0389 Medications before admission Statins 8¢0 7¢1 0¢0352 11¢6 10¢3 0¢0443 ACE inhibitors/ ARBs 34¢2 28¢8 0¢1166 41¢0 35¢2 0¢1196 Beta blockers 20¢7 17¢4 0¢0843 25¢8 21¢4 0¢1033 Angiographic findings Multivessel disease 38¢4 41¢5 0¢0630 46¢3 43¢0 0¢0670 Outcome STEMI 65¢3 74¢7 0¢2061* 65¢2 71¢0 0¢1255x Odds Ratio (95% CI) 0¢64 (0¢520¢79) 0¢2061** 0¢76 (0¢670¢87) 0¢1255xx Data are percentages or means § Standard deviation (SD) unless stated otherwise. *P-value for STEMI for aspirin users versus non-users 0¢0001; ** P-value 0¢0001. x P-value for STEMI for aspirin users versus non-users 0¢0002. xx P-value 0¢0001. Abbreviations: ACE, angiotensin converting enzyme; ARBs, angiotensin receptor blockers; BMI, Body Mass Index; COPD, chronic obstructive pulmonary disease; STEMI, ST elevation myocardial infarction¢. Table 5 Inverse probability of treatment weighting: outcomes sorted by 10-year CVD risk* and aspirin use before index event. Characteristics <10% of risk 10% of risk Aspirin users Aspirin nonusers Standardized Aspirin users N = 1261 Aspirin nonusers Standardized N = 245 N = 2735 difference N = 7882 difference Cardiovascular risk factors History of hypertension 55¢4 51¢6 0¢0776 73¢4 69¢8 0¢0802 History of 41¢1 34¢2 0¢1424 43¢0 40¢2 0¢0558 hypercholesterolemia Former smokers 9¢9 8¢0 0¢0680 8¢0 7¢2 0¢0277 Clinical history COPD 4¢4 2¢9 0¢0811 6¢7 5¢8 0¢0370 Chronic kidney disease 4¢1 3¢3 0¢0405 6¢4 5¢7 0¢0309 Medications before admission Statins 9¢7 7¢6 0¢0760 10¢5 9¢7 0¢0262 ACE inhibitors/ ARBs 31¢4 24¢1 0¢1858 40¢1 36¢0 0¢0852 Beta blockers 20¢3 16¢0 0¢1113 24¢4 21¢3 0¢0723 Angiographic findings Multivessel disease 32¢4 32¢3 0¢0025 47¢7 45¢9 0¢0371 Outcome STEMI 72¢7 75¢1 0¢0540* 62¢3 71¢3 0¢1919x Odds Ratio (95 CI) 0¢88 (0¢661¢19) 0¢0540** 0¢67 (0¢590¢75) 0¢1919xx Data are percentages or means § Standard deviation (SD) unless stated otherwise. Age, diabetes and current smokers were not included in the model as they were represented in the Pooled Cohort Equation¢. *P-value for STEMI for aspirin users versus non-users 0¢4255; ** P-value 0¢4128. x P-value for STEMI for aspirin users versus non-users <0¢0001. xx P-value <0¢0001. Abbreviations: ACE, angiotensin converting enzyme; ARBs, angiotensin receptor blockers; COPD, chronic obstructive pulmonary disease; CVD, cardiovascular disease; STEMI, ST elevation myocardial infarction¢. *10-year CVD risk calculated using the simplified Pooled Cohort Equations¢. risks from the two subgroups were not significantly different from Representation of the main results of the study in diabetic patients is each other (interaction tests: p = 0¢0864). There was thus good evi- illustrated in Fig. 4. dence to support a similar treatment effect between the two sub- groups, which means that aspirin may be effective in those diabetic 4. Discussion patients who concomitantly present with history of hypercholester- olemia and hypertension. Calculations of the interaction tests The current study examined the use of low-dose aspirin and its reported in Figs. 13 are showed in Supplemental Tables 1930. association with the incidence of severe clinical manifestation of CHD R. Bugiardini et al. / EClinicalMedicine 27 (2020) 100548 7 Fig. 2. Estimated effects of aspirin on STEMI in patients with diabetes: distribution by one more CV risk factor. Association between use of aspirin before index event and incidence of ST elevation myocardial infarction in patients with diabetes sorted by its combination with one more traditional risk factor. Abbreviations: OR, Odds Ratio; CI, Confidence Inter- val; CV, cardiovascular; STEMI, ST elevation myocardial infarction. in a large cohort of Caucasian adults without evidence of CV before 0¢661¢19). Nevertheless, some groups of subjects varied in their the qualifying event. The main average treatment effect in the overall response to aspirin. In other words, there was a strong heterogeneity population was that prior aspirin users were less likely to present of treatment effects. When data were sorted by categories of risk, the with STEMI regardless of age and sex with an absolute event reduc- beneficial effect of aspirin was found to be significant in individuals tion of 6¢8% (OR: 0¢73; 95% CI, 0¢650¢82). The results of our work do who were smokers and in those with history of hypertension and not conflict with USPSTF guidelines as we demonstrated that aspirin hypercholesterolemia, but not in adults with diabetes for whom aspi- treatment reduced the risk of STEMI in the higher CV risk group rin is still at present recommended in patients with 10% or more 10- (absolute difference 9¢0%; OR, 0¢67; 95%CI 0¢590¢75), but not in the year CV risk [9,14]. Notably, a large proportion of subjects in our lower risk group (absolute difference 2¢4%; OR ratio, 0¢88; 95%CI study were taking statins and antihypertensive drugs. Medications Fig. 3. Estimated effects of aspirin on STEMI in patients with diabetes: distribution by two or more CV risk factors. Association between use of aspirin before index event and inci- dence of ST elevation myocardial infarction in patients with diabetes sorted its combination with two or more traditional risk factors. Abbreviations: OR, Odds Ratio; CI, Confidence Interval; CV, cardiovascular; STEMI, ST elevation myocardial infarction. 8 R. Bugiardini et al. / EClinicalMedicine 27 (2020) 100548 Fig. 4. Aspirin and prevention of STEMI in diabetes. Abbreviations: STEMI, ST elevation myocardial infarction. were well balanced between aspirin users and nonusers by inverse prevention trial in diabetes has been unpromising [3]. In the ASCEND, probability of treatment weighting analyses. Thus, one could reason- over 15,000 middle-aged or older diabetic patients were randomized ably conclude that our study examined the incremental benefit of to daily aspirin or placebo. During an average follow-up of approxi- aspirin, added to other standard preventive interventions. These find- mately 7 years, the incidence of serious vascular events was modest ings deserve some more considerations. with one percentage point lower in the aspirin group than in the pla- STEMI remains the most significant contributor to morbidity and cebo group. The risk difference was seen mainly in the rate of tran- mortality worldwide, despite a declining incidence and better sur- sient ischemic attacks. On the opposite, the trial showed no vival rates [15]. It is not possible, however, to know exactly why one significant effect of aspirin use, as compared with placebo, on the person dies from STEMI and another does not, as mortality from rates of myocardial infarction and vascular death. These data are con- STEMI can be related to multiple factors including age, Killip class, sistent with our findings on aspirin use for prevention of STEMI. Non- time delay to treatment, renal failure, number of diseased coronary diabetic patients demonstrated large benefits from aspirin (OR: 0¢64; arteries, and left ventricular ejection fraction. Mortality data has to 95%CI: 0¢560¢73), yet diabetic patients failed to do so (OR, 1¢10; 95% be interpreted cautiously at individual level. Yet, STEMI denotes a CI, 0¢891¢35). So, what is the role of aspirin in primary prevention of high risk of mortality at population-level. The mortality rate of STEMI CV disease in patients with diabetes? in our cohort was approximately twice the mortality of NSTE-ACS The scene is familiar in primary care practice. Most diabetic (OR: 1¢93; 95%CI: 1¢59  2¢35). Thus, our data support the high-risk patients have at least one or two more risk factors for CHD. What to nature of STEMI and its potential role in risk-stratification algo- do, therefore, with a 65-year-old man with diabetes and hyperten- rithms. sion treated with metformin and ACE inhibitors? Whether aspirin Early randomized evidence, suggested differences in response to may have a margin of benefit in such a patient is still unknown. To aspirin for primary prevention between diabetic and nondiabetic address this concern, we performed a sequential switchover design subjects [1618]. As well, the evidence from the most recent primary [19]. Diabetes served as its own control balancing the effects of the R. Bugiardini et al. / EClinicalMedicine 27 (2020) 100548 9 other risk factors by inverse probability of treatment weighting. The which may contribute to potentiate the extra-platelet COX-1 -related same modeling approach was used to estimate the effect of diabetes metabolic pathways of CV risk. So, it is not surprising that smoking in combination with one or multiple risk factors. The first model con- may contribute to exceed the threshold of risk at which aspirin is sisted of two linked risk factors, of which one was diabetes. The sec- assumed to become beneficial in terms of cardiovascular disease pre- ond model comprised diabetes and two of the remaining three risk vention. factors under scrutiny. Finally, a single model structure, which New findings by our study include the potential of aspirin to reduce includes all four conventional risk factors was performed. Different the risk of STEMI in diabetic patients who concurrently present with answers were derived in diabetic patients depending on whether the hypertension and hypercholesterolemia. Hypertension and hypercho- associated risk factor was hypertension, hypercholesterolemia, smok- lesterolemia are important risk factors for the development of new ing or a combination of such factors. No differential effect of aspirin CHD events in patients with diabetes [31]. Hypertension and hyper- was seen in diabetic patients who just had one additional major risk cholesterolemia are also known to increase platelet activation. An factor. Low-dose aspirin was insufficient to confer protection from increased level of circulating monocyte-platelet aggregates represents STEMI in diabetic patients who were smokers. Conversely, preven- one of the most robust markers of platelet activation in hypertension tion of STEMI through aspirin use was effective in diabetics when [32]. Hyperlipidemia is associated with shortened platelet survival and these patients were concomitantly affected by hypertension and high platelet production [33]. Under these circumstances, inhibition hypercholesterolemia, in the absence of smoking. In sum, our study by aspirin of COX-1 may avoid continued production of new uninhib- suggests that when making individual treatment decisions in diabe- ited platelets, and circulating monocyte-platelet aggregates. All of tes, clinicians and patients should consider not only the 10-year risk these are assumptions. Further investigations are required to better of CV disease, but also the number and the quality of CV risk factors. define a clearer understanding of the mechanistic effects of aspirin. The coexistence of hypertension and hypercholesterolemia with dia- Subgroup analyses are generally not considered to provide defini- betes is particularly pernicious because of the strong linkage of these tive evidence for several reasons, including the need of a specific conditions with all CV diseases [20]. prior suspicion of the existence of a particular interaction, the statisti- One more source of uncertainty merits attention. The USPSTF rec- cal methods used to identify interactions, and the spurious associa- ommends to calculate 10-year risk models for estimating life-years tions that may arise as a result of multiple testing [34]. Our analysis gained per 10,000 patients taking aspirin, stratified by age and CV risk addressed each of these issues. There were sufficient clinical data factors [9]. However, the 10-year CV risk model accuracy is controver- concerning aspirin-based heterogeneity of treatment effects in dia- sial [21,22] as it incorporates decision modeling that might not reflect betics [3,1618]. We identified the interaction between diabetes and the incremental benefit of aspirin in contemporary patients who low aspirin therapy using tests of interaction on the log scale [13]. already are addressing CV risk through pharmacologic and lifestyle We addressed concern about multiple testing by matching patients measures. The contemporary approach in risk factor control might using inverse probability of treatment weighting [10]. The magnitude potentially reduce the benefits of aspirin. The ASCEND trial did not of the interaction between diabetes and aspirin prophylaxis and the provide sufficient information to address this question. In the ASCEND, precision in its estimation in all clinical settings lead us to believe hypercholesterolemia was managed according to the current standard that the interaction we identified is clinically relevant and probably of care, and as so, a large proportion of the participants were random- not a statistical artefact. ized with statins [3]. The trial, instead, did not notice the use of antihy- Our findings are particularly important for practice. It is common pertensive agents, which may as well be of relevance for outcomes. occurrence that diabetic patients come into a physician’s office for a Differences between antihypertensive drugs exist with respect to tar- routine visit. Aspirin is still often on their medication list. Patients get-organ damage and prevention of CV events. In a subgroup analysis should, therefore, make an informed decision to discontinue or retain of diabetic patients in the LIFE (Losartan Intervention For Endpoint their aspirin. Physicians pull up the 2019 ACC/AHA guidelines [35], Reduction) study [23], treatment with a strategy based on an ARB sig- the 2016 USPSTF recommendations [9] and the 2019 American Dia- nificantly reduced CV morbidity and mortality compared with treat- betes Association (ADA) statement [14] on aspirin online, to discuss ment with beta-blockers. More recently, the use of ACE inhibitors was these indications together. The USPSTF guidelines suggested that the associated with a significant reduction in CV-related mortality in a presence of diabetes did not alter the effectiveness of aspirin therapy broad spectrum of patients with diabetes and no CV disease [24]. This in reducing CV disease events, and recommended low-dose aspirin issue is further compounded by the fact that beta-blocker use in diabe- for adults aged 50 to 69 years whose 10-year CV risk exceeded 10%. tes may be associated with an increased risk for cardiovascular events The ACC/AHA guidelines removed the specific 10-year CV risk thresh- [25]. Although FDA guidelines [26] have always called for an analysis old as an inclusion criterion for aspirin consideration and recom- of the impact of concomitant medications in clinical studies, this has mended that low-dose aspirin should be used infrequently in the not always been pursued. Our study did so. In the matched population, primary prevention of CV disease. The ADA Scientific Statement nar- the magnitude of the effects of statins, ACE inhibitors, ARBs and beta- rowed the recommendations in diabetic patients to those with a blockers on the outcome was comparable in aspirin users versus non- 10year CV risk 10%, provided that these patients are aged more users. Thus, our results are unlikely to be caused by one drug interfer- than 50 and less than 70 years and have at least one additional major ing with the effects of the other. risk factor. In this conundrum of scientific opinions, almost all To date, there is no objective estimate about the threshold of risk patients would prefer to defer decisions to their physicians. Yet, to warrant aspirin prophylaxis in diabetes. The entire coagulation physicians may fail to provide patients with basic information about cascade is dysfunctional in diabetes. Increased levels of fibrinogen the risk benefit ratio of such intervention. Assessing bleeding risk and plasminogen activator inhibitor 1 favor both thrombosis and remains a challenge. Better estimation of future ischemic risk is defective dissolution of clots once formed [27]. Platelets adhere to therefore essential to better inform shared decision making with vascular endothelium and aggregate more readily than those in patients. In the present study, we take a step in helping to account healthy individuals [28]. Aspirin mainly inhibits platelet aggregation for patient-specific factors. The amount of benefit of preventing by irreversible acetylation of the cyclo-oxygenase-1 (COX-1) enzyme, STEMI may clearly outweighs the amount of bleeding risk. Diabetes resulting in almost complete inhibition of thromboxane production should be endorsed as a significant part of risk estimates. Modest [29]. Thus, all of these afore mentioned abnormalities in diabetes can- average effects in recent studies may reflect a mixture of considerable not be counteracted by aspirin alone. To some extent, the effects of benefits for some subjects or little benefit, or any, for others. smoking and diabetes are similar. According to the 2014 Surgeon There are a number of strengths to this study. The study tackles an General’s Report [30], smoking may decrease insulin sensitivity, important question to clinicians: the appropriateness and potential 10 R. Bugiardini et al. / EClinicalMedicine 27 (2020) 100548 benefit of aspirin for primary prevention of cardiovascular disease. Declaration of Competing Interest The dataset is large with good ascertainment of cardiovascular risk factors covering a contemporary period of time. The methods are rea- Professor Badimon reports other from Bayer, personal fees and sonable for addressing measured confounding. A further strength other from International Aspirin Foundation, UK, during the conduct was the use of subgroups related to diabetes and other conditions. of the study; other from SANOFI, personal fees from LILLY, grants The study thoroughly took different variations into account alongside from ASTRAZENECA, personal fees from ASTRAZENECA, other from having diabetes, which is relevant to the community and patients. Glycardial, personal fees from BMS/Pfizer, personal fees from PACE, Our study has some potential limitations. First, some of the risk personal fees and other from FICYE (FORUM TO STUDY BEER & LIFE- factors were ascertained by the general practitioner, which might STYLE), outside the submitted work; In addition, Professor Badimon have led to errors in the dataset. This was the best way to handle this has a patent APOj-Gly licensed, a patent IV_STATIN pending, and a issue. Blood pressure value after an ACS is potentially confounded patent DJ1-F pending. All other authors have nothing to report. because it might have fallen in some subjects as a result of heart fail- ure. As well, stress hyperglycemia may frequently occur and is not an Funding indication of disease. Although we acknowledge some misclassifica- tions, it is unlikely that these misclassifications differentially affect None. diabetic over nondiabetic patients and, thus, are unlikely to modify the main difference that we found. Another limitation is that we were unable to assess the exact duration of aspirin therapy. Out- Supplementary materials comes may be dependent on frequency of aspirin exposure. We relied on what the patients reported at hospital admission. This could result Supplementary material associated with this article can be found in possible misclassification effects of unknown size and direction. in the online version at doi:10.1016/j.eclinm.2020.100548. Further, a general problem of the approach chosen in the present study is that analyses of bleeding events of low-dose aspirin cannot References be ascertained. Many questions remain regarding the decision as to which patients must weigh the benefits of chronic aspirin therapy [1] O’Brien CW, Juraschek SP, Wee CC. Prevalence of aspirin use for primary preven- tion of cardiovascular disease in the United States: results from the 2017 National against the possible risks associated with its use, including the risk of Health Interview Survey. Ann Intern Med 2019;171(8):596–8. intracerebral and subarachnoid hemorrhage. In the ASCEND trial the [2] Gaziano JM, Brotons C, Coppolecchia R, et al. Use of aspirin to reduce risk of initial incidence of major bleeding events was 1 percentage point higher vascular events in patients at moderate risk of cardiovascular disease (ARRIVE): a randomised, double-blind, placebo-controlled trial. Lancet 2018;392 with aspirin than with placebo (4.1% vs. 3.2%). The incidence of hem- (10152):1036–46. orrhagic stroke was similar among persons in the aspirin group and [3] Bowman L, Mafham M, Wallendszus K, et al. Effects of aspirin for primary preven- among those in the placebo group (0.3.vs 0.3%) [3]. Based on the rar- tion in persons with diabetes mellitus. N Engl J Med 2018;379(16):1529–39. ity of hemorrhagic stroke risk, concerns about this risk should not [4] Yeh RW, Sidney S, Chandra M, Sorel M, Selby JV, Go AS. Population trends in the incidence and outcomes of acute myocardial infarction. N Engl J Med 2010;362 discourage appropriate patients from using low-dose aspirin. Accord- (23):2155–65. ingly, our study demonstrates that aspirin has a high margin of bene- [5] McNeil JJ, Wolfe R, Woods RL, et al. Effect of aspirin on cardiovascular events and fit as it substantially reduces the rates of STEMI in specific higher-risk bleeding in the healthy elderly. New England Journal of Medicine 2018;379 (16):1509–18. comorbid conditions such as in hypertension, hypercholesterolemia [6] Ridker PM, Cook NR, Lee IM, et al. A randomized trial of low-dose aspirin in the and smoking. The presence of diabetes in combination with other primary prevention of cardiovascular disease in women. N Engl J Med 2005;352 risk factors should not necessarily be viewed as a guarantee of a fail- (13):1293–304. [7] Cenko E, Yoon J, Kedev S, et al. Sex differences in outcomes after STEMI: ure of aspirin prophylaxis. effect modification by treatment strategy and age. JAMA Intern Med 2018;178 In conclusion, our study underscores the importance of investiga- (5):632–9. tions on the heterogeneity of treatment efficacy. The observed interac- [8] Cenko E, van der Schaar M, Yoon J, et al. Sex-related differences in heart failure after ST-segment elevation myocardial infarction. J Am Coll Cardiol 2019;74 tions between presence of diabetes and its combination with traditional (19):2379–89. risk factors and aspirin prophylaxis can be confirmed only by a very [9] Bibbins-Domingo K. Aspirin use for the primary prevention of cardiovascular dis- large number of risk -stratified, randomized, controlled trials of low ease and colorectal cancer: U.S. preventive services task force recommendation statement. Ann Intern Med 2016;164(12):836–45. dose aspirin therapy. In the absence of definitive evidence from trials, [10] Austin PC, Stuart EA. Moving towards best practice when using inverse probabil- we believe that our data provide sufficient grounds for a reexamination ity of treatment weighting (IPTW) using the propensity score to estimate causal of the use of aspirin therapy for primary prevention of CV disease. treatment effects in observational studies. Stat Med 2015;34(28):3661–79. [11] Beretta L, Santaniello A. Nearest neighbor imputation algorithms: a critical evalu- ation. BMC Med Inform Decis Mak 2016;16 Suppl 3:74. Author contributions [12] Troyanskaya O, Cantor M, Sherlock G, et al. Missing value estimation methods for DNA microarrays. Bioinformatics 2001;17(6):520–5. RB contributed to study design, data interpretation, literature [13] Altman DG, Bland JM. Interaction revisited: the difference between two esti- mates. BMJ 2003;326(7382):219. search and writing and editing of the manuscript. SP contributed to [14] Association American Diabetes. Cardiovascular disease and risk management: study design and editing of the manuscript. JY and MvdS contributed standards of medical care in diabetes-2019. Diabetes Care 2019:S103–S23. to data analysis, and data interpretation. SK, MV, ZV contributed to [15] As Orlandini, R Dí az, Wojdyla D, et al. Outcomes of patients in clinical trials with ST-segment elevation myocardial infarction among countries with different gross data collection and editing of the manuscript. MB contributed to the national incomes. Eur. Heart J. 2006;27(5):527–33. literature search and editing of the manuscript. DM contributed to [16] Kassoff A, Buzney SM, McMeel JW, et al. Aspirin effects on mortality and morbid- data collection and editing of the manuscript. OM contributed to data ity in patients with diabetes mellitus: early treatment diabetic retinopathy study report 14. JAMA 1992;268(10):1292–300. interpretation and editing of the manuscript. EC contributed to study [17] Belch J, MacCuish A, Campbell I, et al. The prevention of progression of arterial design, data interpretation and critical revision of the manuscript. LB disease and diabetes (POPADAD) trial: factorial randomised placebo controlled contributed to data interpretation and editing of the manuscript. trial of aspirin and antioxidants in patients with diabetes and asymptomatic peripheral arterial disease. BMJ 2008;337 a1840-a. [18] Ogawa H, Nakayama M, Morimoto T, et al. Low-dose aspirin for primary preven- Data sharing statement tion of atherosclerotic events in patients with type 2 diabetes: a randomized con- trolled trial. JAMA 2008;300(18):2134–41. The source codes of the study are uploaded on https://github.com/ [19] Maclure M, Mittleman MA. Should we use a case-crossover design. Annu Rev Pub- lic Health 2000;21:193–221. jsyoon0823/Treatment_Phenotype. They will be available to anyone [20] Wilson PW, Kannel WB, Silbershatz H, D'Agostino RB. Clustering of metabolic fac- who wishes to access these data. tors and coronary heart disease. Arch Intern Med 1999;159(10):1104–9. R. Bugiardini et al. / EClinicalMedicine 27 (2020) 100548 11 [21] Brett A.S. Accuracy of the ACC/AHA cardiovascular risk calculator is challenged. [28] Harrison HE, Reece AH, Johnson M. Effect of insulin treatment on prostacyclin in 2015. https://www.jwatch.org/na37408/2015/03/25/accuracy-acc-aha-cardio- experimental diabetes. Diabetologia 1980;18(1):65–8. vascular-risk-calculator-challenged2020). [29] Patrono C. The multifaceted clinical readouts of platelet inhibition by low-dose [22] DeFilippis AP, Young R, Carrubba CJ, et al. An analysis of calibration and discrimi- aspirin. J Am Coll Cardiol 2015;66(1):74. nation among multiple cardiovascular risk scores in a modern multiethnic cohort. [30] U.S. Department of Health and Human Services (USDHHS). The Health Conse- Ann Intern Med 2015;162(4):266–75. quences of Smoking: 50 Years of Progress. A Report of the Surgeon General. [23] Lindholm LH, Ibsen H, Dahlo € f B, et al. Cardiovascular morbidity and mortality in Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease patients with diabetes in the Losartan Intervention For Endpoint reduction in Control and Prevention, National Center for Chronic Disease Prevention and hypertension study (LIFE): a randomised trial against atenolol. Lancet 2002;359 Health Promotion, Office on Smoking and Health; 2014 (9311):1004–10. [31] Tight blood pressure control and risk of macrovascular and microvascular compli- [24] Eurich DT, Majumdar SR, Tsuyuki RT, Johnson JA. Reduced mortality associated cations in type 2 diabetes: UKPDS 38. UK prospective diabetes study group. BMJ with the use of ACE inhibitors in patients with type 2 diabetes. Diabetes Care 1998;317(7160):703–13. 2004;27(6):1330–4. [32] Gkaliagkousi E, Ritter J, Ferro A. Platelet-derived nitric oxide signaling and regula- [25] Tsujimoto T, Sugiyama T, Shapiro MF, Noda M, Kajio H. Risk of cardiovascular tion. Circ Res 2007;101(7):654–62. events in patients with diabetes mellitus on b-blockers. Hypertension 2017;70 [33] Harker LA, Hazzard W. Platelet kinetic studies in patients with hyperlipoproteine- (1):103–10. mia: effects of clofibrate therapy. Circulation 1979;60(3):492–6. [26] FDA. E3 structure and content of clinical study reports. 1996. https://www.fda. [34] Oxman AD, Guyatt GH. A consumer's guide to subgroup analyses. Ann Intern Med gov/regulatory-information/search-fda-guidance-documents/e3-structure-and- 1992;116(1):78–84. content-clinical-study-reports. [35] Arnett Donna K, Blumenthal Roger S, Albert Michelle A, et al. 2019 ACC/AHA [27] Festa A, D’Agostino R, Mykk€anen L, et al. Relative contribution of insulin and its guideline on the primary prevention of cardiovascular disease: a Report of the precursors to fibrinogen and PAI-1 in a large population with different states of American College of Cardiology/American Heart Association Task Force on Clini- glucose tolerance. Arterioscler Thromb Vasc Biol 1999;19(3):562–8. cal Practice Guidelines. Circulation 2019;140(11):e596–646.